University of Groningen Control of metabolic flux by ... · PDF fileControl of Metabolic Flux...

University of Groningen

Control of metabolic flux by nutrient sensorsOosterveer, Maaike

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2009

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Oosterveer, M. H. (2009). Control of metabolic flux by nutrient sensors Groningen: s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 05-05-2018

https://www.rug.nl/research/portal/en/publications/control-of-metabolic-flux-by-nutrient-sensors(4c97f760-ebb2-411f-93c4-0676b11f30f8).html

Control of Metabolic Fluxby Nutrient Sensors

Maaike Hélène Oosterveer

Paranimfen Marijke Schreurs Anniek Koolman

Research described in this dissertation was funded by the Dutch Diabetes Research Foundation, grant 2002.00.041. Printing of this dissertation is fi nancially supported by Diabetes Fonds NederlandGroningen University Institute for Drug ExplorationJ.E. Jurriaanse Stichting Nederlandse Vereniging voor Hepatologie Rijksuniversiteit Groningen, Faculteit der Medische Wetenschappen

AuthorMaaike H. OosterveerCopyright© 2009. All rights reserved.No part of this publication may be reproduced or transmitted in any form by any means, electronic or mechanical, including photocopy, recording or any information storage and retrieval system, without written permission of the author.

Cover photographyThijs Schouten Fotografi e, www.thijsschouten.com

Lay-out and cover designEel & Fishwive Productions Ltd.

PrintingGildeprint Drukkerijen

RIJKSUNIVERSITEIT GRONINGEN

Control of Metabolic Fluxby Nutrient Sensors

Proefschrift

ter verkrijging van het doctoraat in deMedische Wetenschappen

aan de Rijksuniversiteit Groningenop gezag van de

Rector Magnifi cus, dr. F. Zwarts,in het openbaar te verdedigen op

woensdag 21 oktober 2009om 14.45 uur

door

Maaike Hélène Oosterveer

geboren op 10 juli 1980te Nijmegen

Promotor Prof. dr. F. Kuipers

Copromotor Dr. D-J. Reijngoud

Beoordelingscommissie Prof. dr. K.N. FraynProf. dr. A.K. GroenProf. dr. J.A. Romijn

ISBN 978-90-367-3943-6 (printed)978-90-367-3944-3 (digital)

‘Zolang ik kan blijven zweven

tussen vraag en antwoord door

houd ik mijn leven op de rails

maar zit ik nergens in het spoor’Uit: Grijs, Veldhuis en Kemper

Contents

1 - Introduction 9

2 - Lxrα defi ciency hampers the hepatic adaptive response to fasting in mice 25

3 - An increased fl ux through the glucose-6-phosphate pool in enterocytes delays glucose absorption in Fxr -/- mice

41

4 - PPARα activation simultaneously induces hepatic fatty acid oxidation, synthesis and elongation in mice

57

5 - Fish oil potentiates high fat diet-induced peripheral insulin resistance in mice

75

6 - High-fat feeding induces hepatic fatty acid elongation in mice 91

7 - General Discussion 107

Frequently used Abbreviations

Supplemental Material

References

Summary

Samenvatting

Dankwoord

Biografi e/Biography

List of Publications

1 Introduction

METABOLIC FLUX CONTROLComplex organisms possess multiple ‘nutrient sensing’ systems that allow adequate biochemical ad-

aptations to ensure suffi cient energy supply to organs and tissues and to facilitate the storage of su-

perfl uous nutrients. The effi cacy of these systems is illustrated by the body’s ability to cope with the

diurnal alternation between feeding and fasting, and with excessive energy demands during exer-

cise. Oversupply of energy and nutritional dysbalance trigger adaptive physiology and interfere with

regulatory pathways. As a consequence, these conditions predispose to development of metabolic

disturbances such as type 2 diabetes and cardiovascular diseases. Insight into the pathophysiologi-

cal mechanisms is required to defi ne strategies for prevention and treatment of these diseases.

The breakdown, rearrangement and storage of nutrients is accounted for by metabolic fl uxes,

which are defi ned as the fl ow rates of molecules through biochemical pathways. These fl ow rates

are determined by substrate availability and enzyme activities. Cellular nutrient status per se deter-

mines substrate availability. Some nutrients furthermore control enzyme activities via feedforward

control upon substrate binding or via feedback control by downstream metabolites. Changes in

nutrient and/or energy status may also induce sensing systems that exert post-translational modi-

fi cations of regulatory proteins. These modifi cations shift the equilibrium between the active and

inactive state of an enzyme and/or aff ect its stability. Thus, nutritional status determines metabolic

fl ux by both direct and indirect mechanisms.

For decades, hormonal networks have been considered as the major sensing pathways responsi-

ble for indirect fl ux control by nutrients. The recent identifi cation of transcription factors that control

the gene expression of enzymes has added a new level of complexity to the regulation of metabolic

fl uxes. Many of these transcriptional regulators are activated by nutrients or their metabolites. There-

fore, these small-molecule sensors are interesting targets to modulate metabolic fl uxes.

The studies described in this dissertation consider the adaptive mechanisms by which the body

responds to changes in nutrient availability. Glucose, fatty acids and amino acids are the major di-

etary energy suppliers. The work focuses on glucose and fat metabolism, and particularly addresses

the role of transcriptional regulators.

GLUCOSE HOMEOSTASISGlucose is the primary metabolic fuel for complex organisms. Although most cells are able to ex-

hibit substrate switching when glucose supply is limited, some almost exclusively depend on glu-

cose. This is particularly true for brain cells and erythrocytes: their functioning is severely impaired

if glucose concentrations are persistently low. High glucose concentrations, on the other hand, are

cytotoxic and cause tissue damage. Tight control of glycemia is thus required to guarantee optimal

functioning.

Most dietary glucose is ingested as a multimer complex (i.e., carbohydrates). After digestion of

these complexes, glucose molecules are taken up into intestinal cells from where they are trans-

ported into the bloodstream. Intestinal uptake represents the major route for glucose input into

the circulation in the absorptive or postprandial phase (i.e., following the intake of a meal). Glucose

transport across plasma membranes is facilitated by glucose transporters (GLUTs). Once inside the

cell, glucose is phosphorylated to glucose-6-phosphate (G6P) by glucokinase (GK) in hepatocytes

1 1Chapter 1

and pancreatic β-cells and by hexokinases (HK) in all other cell types. G6P has diff erent intracellular

fates. A limited amount of G6P is stored as glycogen in liver and skeletal muscles. Furthermore, G6P

is used for energy supply by its conversion into pyruvate via the glycolytic pathway. Pyruvate kinase

(PK) catalyzes the fi nal step of glycolysis. Pyruvate enters the mitochondria and is used as a substrate

in the tricarboxylic acid (TCA) cycle to generate energy in the form of adenosine triphosphate (ATP).

If ATP supply is suffi cient, pyruvate is used for de novo fatty acid synthesis. The pentose phosphate

pathway (PPP) represents another route for intracellular G6P. Although the PPP involves glucose oxi-

dation, its primary role is anabolic rather than catabolic because it generates reducing equivalents

required for biosynthetic routes.

Blood glucose concentrations decrease as soon as intestinal absorption is completed (postabsorp-

tive phase). In this phase, glucose consumption by most tissues is reduced while glucose production

represents the major route of glucose input into the circulation and secures substrate supply for the

brain. Glucose-6-phosphatase (G6Pase) systems enable glucose production from G6P in liver, kid-

ney and intestine. In the postabsorptive state, the relative contributions of kidney and intestine are

limited [1]. Under these conditions, hepatic G6P is derived from glycogen breakdown via glycogen

phosphorylase (GP). Upon prolonged fasting, de novo synthesis of G6P from 3-carbon precursors, or

gluconeogenesis, is induced. Lactate and alanine are converted into pyruvate, while glycerol can

be used as a gluconeogenic substrate via triose phosphate. G6Pase activity mediates G6P transport

from the cytosol into the endoplasmic reticulum (ER) by glucose-6-phosphate translocase (G6Pt)

and its subsequent hydrolysis to glucose by glucose-6-phosphate hydrolase (G6Ph). Glucose is fi -

nally transported from the ER into the circulation.

FATTY ACID HOMEOSTASISFatty acids represent the second major metabolic substrate. Non-esterifi ed fatty acids (NEFA) are

extremely cytotoxic [2]. Circulating NEFA concentrations show relatively little variation and cellular

fatty acid uptake, transport and storage are heavily regulated [3].

In the postprandial phase, dietary fatty acids mainly enter the body in the form of triglycerides

(TGs), consisting of fatty acids complexed to glycerol. After TG digestion, fatty acids are transported

into intestinal cells where they are re-esterifi ed to form TG or cholesterol-esters (CEs) which are as-

sembled into chylomicrons. These relatively large particles enter the circulation via the lymphatic

system. Chylomicron-associated TGs are hydrolyzed by the action of lipoprotein lipase (LPL). Uptake

of NEFA by tissues and their intracellular traffi cking is mediated by fatty acid transporters and fatty

acid binding proteins [4,5]. Catabolism of dietary long-chain fatty acids by β- oxidation in mitochon-

dria and peroxisomes is mediated by multienzyme complexes [6]. Fatty acids enter mitochondria

and peroxisomes as carnitine complexes, which are transported across the membranes by carnitine

transferases. Surplus fatty acids are re-esterifi ed with glycerol and cholesterol and stored as TGs and

CEs. These processes are mediated by glycerol-3-phosphate acyltransferase (GPAT), diacylglycerol

acyltransferases (DGATs) and acyl-CoA:cholesterol acyltransferases (ACAT), respectively. Fatty acids

can also be synthesized from excess glucose in liver and adipose tissue in the postprandial phase.

This process, which is called de novo lipogenesis, is initiated by transport of citrate across the mito-

chondrial membrane into the cytosol. Here, citrate is converted into acetyl-CoA by ATP citrate lyase

(ACL). Acetyl-CoA is subsequently converted into malonyl-CoA by acetyl-CoA carboxylase (ACC).

1 2 Control of energy homeostasis through metabolic fl uxes

Seven malonyl-CoA molecules are condensed to a singly acetyl-CoA molecule, thereby forming pal-

mitic acid. This process is mediated by the fatty acid synthase (FAS) complex. Specifi c types of fatty

acids are stored as TG and CE while others are required for particular physiological functions, such

as eicosanoid production and phospholipid synthesis. In the ER, diff erent fatty acid elongation and

desaturation enzymes facilitate the synthesis of these fatty acids from newly synthesized palmitic

acid or from diet-derived fatty acids [7,8]. Part of the TGs and CEs synthesized in the liver are pack-

aged into very low density lipoprotein (VLDL) particles, which are released into the circulation. These

particles provide fatty acids to peripheral tissues via LPL-mediated lipolysis.

In the postabsorptive phase, TGs stored in adipose tissue are hydrolyzed into glycerol and NEFA,

which are released into the circulation. Under these conditions, most tissues switch to fatty acid oxi-

dation and circulating NEFAs (complexed to albumin) serve as the major energy substrates. Part of

the acetyl-CoA generated from hepatic β-oxidation is converted into ketone bodies, which provide

an alternative fuel for the brain during prolonged fasting. The liver also plays a role in fatty acid supply

to peripheral tissues via the secretion of VLDL. Glycerol released by adipocytes is used as a 3-carbon

precursor for hepatic gluconeogenesis. Not all NEFA that are taken up by the liver are immediately

oxidized. This results in hepatic TG accumulation upon fasting [9,10]. An overview of cellular glucose

and fatty acid metabolism in the postprandial and postabsorptive phase is given in Figure 1.

1 3Chapter 1

Figure 1. Schematic overview of glucose and fatty acid metabolism in the postprandial and postabsorptive phases.

glucose GLUTGLUTHK/GK

GS GP

glycogen

triose phosphate

PK

pyruvate

acetyl-CoA

TCA cycleTCA cycleTCA cycle

glucose-6-phosphate

PPP

PEP

citrate

acetyl-CoA

malonyl-CoA

fatty acid

ACC

POSTPRANDIAL PHASE

FAS / elongation & desaturation

glucose oxidation

ACL

fat storage

DGAT / GPAT / ACAT TG / CE

glucose G6PT/G6PH

GS GP

glycogen

triose phosphate

pyruvate

PK

acetyl-CoA

TCA cycleTCA cycleTCA cycle

glucose-6-phosphate

PPP

PEP

fatty acids

ketone bodies

POSTABSORPTIVE PHASEfat oxidation

glucose production

FATPFATP


REGULATORY PATHWAYS OF GLUCOSE AND LIPID HOMEOSTASIS Direct regulation by energy and nutrient availability

Cells adjust ATP production to their needs. Regulatory systems that control energy homeostasis

therefore not only sense nutrient availability but also shift the balance between nutrient utiliza-

tion and storage. Many of the metabolic adaptations in the postprandial and postabsorptive phase

coincide with the reciprocal actions of the nutrient sensors insulin and glucagon. Cellular energy

and nutrient status, however, also directly aff ect metabolic fl uxes. Low energy availability activated

adenosine monophosphate kinase (AMPK), which modulates the activity of metabolic enzymes by

phosphorylation. For example, if ATP supply is suffi cient, citrate consumption in the TCA cycle is

inhibited. Citrate is then converted into acetyl-CoA, which is used for fat storage via de novo lipoge-

nesis. Furthermore, glucose and fatty acids exert substrate competition. This phenomenon was fi rst

described by Randle et al. as the glucose-fatty acid cycle [11]. Malonyl-CoA produced from excess

glucose inhibits carnitine palmitoyltransferase-1 (CPT1) activity. This enzyme catalyzes acylcarnitine

transport across the outer mitochondrial membrane and hence inhibits fatty acid oxidation when

the glycolytic fl ux is high. On the other hand, the inhibition of CTP1 activity will be released if glyco-

lysis is low and fatty acid oxidation will consequently increase. β-Oxidation products inhibit pyruvate

dehydrogenase (PDH) action via an increased pyruvate dehydrogenase kinase (PDK) activity. As a

consequence, pyruvate conversion into acetyl-CoA is blocked.

Indirect regulation by hormones

Insulin and glucagon are the major hormones that exert indirect fl ux control in response to changes

in glucose availability. These hormones not only aff ect post-translational modifi cation systems (i.e.,

protein kinase and phosphatase activities) but also regulate enzyme expression at the transcriptio-

nal level. In the postprandial phase, increasing blood glucose concentrations trigger insulin release

by pancreatic β-cells. Circulating insulin binds to its receptor (insulin receptor, IR), which initiates se-

veral complex signalling cascades. The insulin receptor substrate (IRS)/phosphatidylinositol 3-kinase

(PI3K) pathway activates protein kinases, which mediate most of insulin’s metabolic actions relevant

for glucose homeostasis.

These include:

Translocation and fusion of intracellular GLUT4 vesicles to the plasma membrane, thereby pro- -

moting glucose uptake into skeletal muscle, adipose tissue and heart.

Induction of glycogen synthase (GS) activity, which facilitates glycogen storage in liver and skel- -

etal muscles.

Suppression of phosphoenolpyruvate carboxykinase ( - Pepck) expression, which encodes a key

gluconeogenic enzyme.

Induction of sterol regulatory element binding protein 1c ( - Srebp-1c) expression, which encodes

a regulator of fatty acid synthesis in liver and adipose tissue.

Inhibition of lipolytic enzyme activity in adipose tissue, thereby suppressing TG hydrolysis. -

1 5Chapter 1

A decrease in the blood glucose concentration arrests insulin release and its actions. Low blood

glucose concentrations furthermore trigger glucagon secretion by pancreatic α-cells. Glucagon in-

teraction with its receptor ultimately results in:

Suppression of GS and induction of GP activity, thereby promoting glycogen breakdown. -

Induction of - G6ph and Pepck expression, which encode gluconeogenic enzymes.

Suppression of PK activity, thereby inhibiting glycolysis. -

Induction of lipolytic enzyme activity in adipose tissue, thereby promoting TG hydrolysis. -

NUCLEAR RECEPTORS AND TRANSCRIPTION FACTORS: THEIR ROLE IN NUTRITIONAL CONTROL OF METABOLIC FLUXTranscriptional regulators are proteins that control gene expression by binding to specifi c response

elements (REs) located in the promoter sequences of genes. The activity of nuclear receptors and

transcription factors depends on their cellular location and structural conformation. Nuclear recep-

tors represent a superfamily of transcription factors that are mostly ligand-activated. These receptors

share a common structural and functional organization. This consists of a NH2-terminal domain for

ligand-independent transactivation, a DNA-binding domain required for proper targeting to the REs,

a connecting hinge region that allows protein fl exibility and a ligand-binding domain that exerts lig-

and-dependent transactivation. Upon binding, some nuclear receptors are fi rst translocated to the

nucleus upon ligand binding. Nuclear receptors bind to their REs as monomers, but more often as

homodimers or heterodimers. In addition, dephosphorylation and ligand activation of (RE-bound)

receptors induce conformational changes, thereby modulating the affi nity for certain co-repressor

and co-activator proteins. These in turn determine whether a target gene is induced or suppressed.

Nuclear receptor action is depicted in Figure 2.

CR

RE

CA

transcription

LBD

DBD

ligands (nutrients)

CRPP

CRCR

CACACRCRPP

Figure 2. Schematic overview of transcriptional regulation upon ligand-activation of nuclear receptors.


Based on their ligand-binding properties, nuclear receptors can be divided into three classes. Nuclear

hormone receptors represent those that bind hormones with a high affi nity. Well-known examples

are the glucocorticoid receptor and the estrogen receptor. Those for which the ligand still needs to

be identifi ed are called orphan receptors. Several of the recently adopted receptors bind metabolic

substrates and intermediates, but also xenobiotics and drug metabolites. Nutrients and their me-

tabolites exert indirect fl ux control via these regulators. They serve as ligands for certain adopted

receptors and hence modulate their transcriptional activity. Table 1 provides an overview of the dif-

ferent transcriptional regulators, their nutrient sensitivity and metabolic regulation.

Table 1. Overview of nutrient-sensing transcription factors and their metabolic actions.

Transcription factor Sensitive to Metabolic regulation

Ligand-activated nuclear receptors

PPARα fatty acids fatty acid oxidation

PPARβ fatty acids fatty acid oxidation

cholesterol transport

glucose transport

PPARγ fatty acids lipid storage

glucose transport

LXRα/β oxysterols cholesterol transport

bile acid synthesis

fatty acid synthesis

glucose transport

FXRα/β bile acids bile acid synthesis

cholesterol transport

glucose transport

glucose oxidation


Other transcription factors

SREBP-1c cholesterol fatty acid synthesis

ChREBP glucose glucose oxidation


Peroxisome Proliferator Activated Receptors (PPARs)

PPARs are key regulators of lipid homeostasis. PPARs are ligand-activated by fatty acids, in particu-

lar by polyunsaturated fatty acids (PUFA) and eicosanoids [12]. PPARs form heterodimers with the

Retinoid X Receptor (RXR), which is ligand-activated by retinoic acid. There are three PPAR isotypes,

encoded by separate genes.

1 7Chapter 1

PPARα (NR1C1) is highly expressed in liver, brown adipose tissue, heart and skeletal muscle. Upon

activation, PPARα induces the expression of enzymes involved in fatty acid mobilization, uptake,

transport and catabolism. In the fasted state, PPARα is activated by NEFA released from adipose tis-

sue. This facilitates energy supply and enables ketogenesis. Because of this, PPARα is an important

mediator of the adaptive response to fasting [13,14].

Two PPARγ (NR1C3) isoforms exist. PPARγ1 is mainly expressed in adipose tissues, but also in the

colon, spleen, retina, hematopoietic cells and skeletal muscles. PPARγ2, on the other hand, is pre-

dominantly expressed in white and brown adipose tissue. PPARγ coordinates adipocyte diff erentia-

tion and proliferation [15].

PPARβ/δ (NR1C2) is ubiquitously expressed. PPARβ/δ promotes fatty acid oxidation in skeletal

muscle and adipose tissue [16]. In addition, PPARβ/δ is involved in cholesterol export in intestine

and macrophages [17]. All three PPAR isoforms furthermore mediate infl ammatory responses [18].

Liver X Receptors (LXRs)

LXRs are major players in control of cholesterol and fatty acid metabolism [19] and infl ammatory re-

sponses [18]. The two LXR isotypes are ligand-activated by mono-oxidized derivatives of cholesterol.

LXRs also heterodimerize with RXR. LXR binding to its response elements is inhibited by PUFA [20].

LXRα (NR1H3) is highly expressed in liver, and to a lower extent in kidney, intestine, adipose tissue

and macrophages while LXRβ (NR1H2) is ubiquitously expressed. LXR target genes encode enzymes

involved in cholesterol effl ux and disposal, i.e., bile acid synthesis, hepatobiliary transport and fecal

excretion. In addition, LXRs increase fatty acid synthesis, both directly and indirectly via the induc-

tion of Srebp-1c [21].

Farnesoid X Receptors (FXRs)

FXRs, which control bile acid and cholesterol metabolism are ligand-activated by bile acids. FXRs can

either act as monomer, or form heterodimers with RXR. There are two FXR isotypes. FXRα is mainly

expressed in liver and adrenals. FXRβ expression is higher compared to that of FXRα and most domi-

nant in intestine and kidney. FXR activation serves to protect from toxic accumulation of bile acids,

by inhibition of bile acid uptake and synthesis genes while inducing bile acid export systems. FXRs

have also been implicated in the regulation of glucose and fatty acid homeostasis [22].

Sterol Regulatory Element Binding Proteins (SREBPs)

SREBPS are transcription factors that regulate cholesterol and fatty acid metabolism [23]. There are

two SREBP isotypes (SREBP-1/2), which are predominantly present in liver and adipose tissue. SREBPs

are synthesized as 125 kDa precursor proteins anchored in the ER membrane. Maturation of SREBPs

requires the activation of the SREBP cleavage activating protein (SCAP). SCAP is a sensor of the cho-

lesterol content in the ER membrane, where it is retained in the presence of high cholesterol levels

due to its interaction with the INSIG proteins [24]. When the cholesterol content drops, SCAP escorts

SREBPs from the ER to the Golgi apparatus. Here, SREBPs are cleaved by two diff erent proteases. The

mature 68 kDa SREBP proteins are translocated to the nucleus where they bind to the DNA as mono-

mers. This maturation process of SREBPs is depicted in Figure 3.

SREBP-2 is mainly involved in control of cholesterol biosynthesis [25]. There are two SREBP-1 iso-

forms. SREBP-1a expression is relatively low compared to that of SREBP-1c. SREBP-1c regulates the

expression of fatty acid biosynthesis and esterifi cation genes. Furthermore, Srebp-1c expression is


controlled by LXRs and insulin [26,27]. The subsequent increase in fatty acid synthesis is thought to

support cholesterol esterifi cation and thereby to faciliate cholesterol storage upon LXR activation.

Insulin’s induction of Srebp-1c expression on the other hand, enables storage of excess glucose as fat.

Both SREBP-1 and -2 also induce systems that generate reducing equivalents required for cholesterol

and fatty acid synthesis [28]. PUFA arrest SREBP-1 but not SREBP-2 action, by enhancing its decay

and/or inhibition of its maturation process [29,30].

ER

SREBP

SCAP

Golgi

SRE

transcription

SREBP

SREBP

SCAP

SREBP SREBPS1P S2P

low cholesterol

Figure 3. Schematic overview of transcriptional regulation upon sterol-induced activation of SREBPs.

Carbohydrate Responsive Element Binding Protein (ChREBP)

ChREBP promotes storage of glucose as fatty acids. This transcription factor that is mainly expressed

in liver, adipose tissue and kidney, is activated in response to increased glucose availability. Inactive

ChREBP is phosphorylated by protein kinase A and localized in the cytosol. Activation of ChREBP

occurs by a two-step dephosphorylation: the fi rst triggers its nuclear translocation while the second

allows its binding to DNA. The transcriptional activity of ChREBP requires its heterodimerization with

the Max-like protein X (Mlx). ChREBP regulates the expression of glycolytic and lipogenic genes. The

PPP intermediate xylulose-5-phosphate is thought to promote protein phosphatase 2A (PP2A) activ-

ity, which in turn dephosphorylates ChREBP, thereby increasing its activity [31]. On the other hand,

AMPK [32] and PUFA suppress ChREBP activity by inhibition of its nuclear translocation [33]. This

2-step activation of ChREBP is depicted in Figure 4.

1 9Chapter 1

PPChREBP

PP PPPPChREBPChREBP

PP

glucose GLUTGLUT

triose phosphate

pyruvate

glucose-6-phosphate

PPPPP

ChREBPPP

ChREBPPPPP

ChREBPChREBPPPPP

ChREBP

ChoRE

transcription

ChREBPMlx

ChoRE

ChREBPChREBPMlxMlx

PPChREBP

PPChREBP

PPPPChREBPChREBP

PPPPChREBP

PP2A

Figure 4. Schematic overview of transcriptional regulation upon glucose-induced activation of ChREBP.

DETERMINATION OF METABOLIC FLUXES UPON INTRODUCTION OF STABLE ISOTOPES IN VIVOAs stated earlier, metabolic fl ux is determined by substrate availability and enzyme activities. Sub-

strate concentrations can be assessed by biochemical analysis (metabolomics). This only provides a

static measure of metabolite status: insight into the origin of a substrate pool (i.e., the contribution

of input versus output) is lacking. Information on the actions of a specifi c enzyme can be derived

from analysis of gene expression level (genomics), its cellular abundance (proteomics), and by deter-

mination of its (maximal) activity ex vivo. However, these analyses do not necessarily refl ect the true

activity under physiological conditions.

Fluxomics allows realtime assessment of substrate fl ow in vivo [34,35]. Such measurements can be

performed in isolated cells, perfused organs or intact organisms, thereby providing detailed infor-

mation of metabolic processes from a single cell to complex whole-body organ interplay. Fluxomics

therefore enables the identifi cation and evaluation of (supposed) critical or rate-limiting steps in a

physiological relevant manner. Most commonly used fl uxomics procedures are based on isotopic

labeling. Labeled molecules are introduced into the system and assumed to be metabolized in a

similar manner as those endogenously present. Fluxes are consequently quantifi ed by assessment of

the degree of labeling in the metabolite of interest within a certain timeframe. Secreted or circulat-

ing metabolites can be studied in a dynamic manner by taking serial samples over time. In the past

decades, stably labeled compounds have been proven an excellent alternative for the traditional


radioisotopes. The advantage of stable isotopes it their safe application in human studies. The mass

isotopomer abundances are determined by gas chromatography and mass spectrometry (GC-MS).

The mass isotopomer distribution in turn allows realtime assessment of biosynthetic fl uxes in vivo.

Mass isotopomer distribution analysis (MIDA) represents one of the mathematical approaches to

quantify these fl uxes. This method will be discussed in detail below.

The turnover or rate of appearance (Ra) of a metabolite is derived from the dilution of a labeled

form of this metabolite that is introduced. If physiology and label enrichment are constant (i.e., in

steady-state), the turnover represents in- and output of the metabolite, thus Ra equals the rate of

disappearance (or disposal, Rd). Such measurements are for instance applied to determine whole-

body glucose production and disposal in vivo following introduction of 13C-labeled glucose. The

contribution of diff erent anabolic routes and label recycling are however not accounted for by this

methodology. Accurate assessment of the fl ux through biosynthetic pathways therefore requires a

more sophisticated approach. This is provided by MIDA, introduced by Hellerstein and Neese [36,37].

MIDA enables the analysis of biopolymer synthesis from repetitive addition of monomeric precur-

sors. The precursor pool is isotopically enriched by the introduction of labeled precursor. The isotope

distribution of the synthesized polymer is conform to binominal expansion and depends on the

enrichment of the precursor pool (p) and the number of monomers in the polymer (n). The relation-

ship between the diff erent isotopomers of the polymer is uniquely determined by p, and therefore

insensitive to dilution by unlabeled polymers [36,38]. Stable isotopes are naturally abundant (~1%)

and the measured isotope abundance must therefore be corrected to obtain the excess isotope en-

richment due to label incorporation [39]. The theoretical undiluted isotopomer abundance is subse-

quently calculated at the specifi c p and n. Its dilution (i.e., the relative excess isotopomer abundance)

fi nally represents the fraction of the polymer pool that is newly synthesized (f ). The introduction of 13C-acetate allows the assessment of fractional fatty acid and cholesterol synthesis. In addition, we

have developed methods to determine individual fl uxes of hepatic glucose metabolism upon intro-

duction of 13C-glucose, 13C-glycerol and 2H-galactose [40,41].

2 1Chapter 1

SCOPE AND OUTLINEObesity, insulin resistance and hepatic steatosis represent three components of the metabolic syn-

drome that are typically associated with energy oversupply and nutritional dysbalance. Obesity and

insulin resistance are furthermore characterized by an impaired capability to balance nutrient avail-

ability and substrate utilization [42]. Transcription factors sense nutrient availability and control the

expression of genes encoding metabolic enzymes. Impaired action and overactivity of transcription

factors are associated with metabolic disturbances [21,41,43–48]. Therefore, metabolic abnormali-

ties can be corrected by modulation of transcriptional activity. There are several examples of drugs

that are used to treat disturbances in lipid and glucose metabolism via the action of nuclear re-

ceptors. Fibrates are pharmacological PPARα agonists that are widely used to treat dyslipidemia in

humans [49]. Pharmacological PPARγ activation by thiazolidinedione (TZD) treatment lowers blood

glucose concentrations in insulin-resistant subjects [50]. Most transcription factors are however ex-

pressed in multiple tissues and global targeting may therefore result in undesirable side-eff ects.

For example, pharmacological LXR agonists are potential anti-atherosclerotic drugs because they

reduce cholesterol accumulation in macrophages. These compounds however also induce hepatic

steatosis and the secretion of large VLDL particles [51]. Furthermore, gene expression manipulations

per se will not always result in altered metabolic fl ux, and biochemical changes refl ect a shift in the

balance between anabolic and catabolic processes. Finally, physiological systems are interrelated.

The induction or suppression of a certain a metabolic pathway will therefore aff ect the fl ux through

another route [52].

Altogether, insight into tissue-specifi c actions of transcriptional regulators as well as the whole-

body consequences for intermediary metabolism are required to defi ne optimal strategies to treat

and prevent metabolic diseases. By modulating the activity of several nuclear receptors, we studied

their role in control of metabolic fl uxes relevant for glucose and lipid homeostasis. In Chapter 2, we

investigated the role of LXRα in the control of hepatic carbohydrate metabolism during the feading-

to-fasting transition. We furthermore assessed the physiological relevance of the postulated hepatic

glucose-sensing function of LXR [53]. Chapter 3 focuses on the regulatory action of FXR in glucose

transport across enterocytes. The consequences of pharmacological PPARα activation for hepatic

carbohydrate and lipid metabolism were determined in Chapter 4. We also evaluated the metabolic

adaptations in response to chronic dietary fat oversupply in mice. Therefore, we used two diff erent

high-fat diets. The fi rst was based on beef fat, while in the other diet this fat was partially replaced by

fi sh oil. The consequences for whole-body glucose metabolism and hepatic fatty acid synthesis are

described in Chapter 5 and 6, respectively. We also assessed the eff ects of these dietary interven-

tions on substrate utilization and energy expenditure (Chapter 5).


2 3Chapter 1

M.H. Oosterveer

T.H. van Dijk

A. Grefhorst

V.W. Bloks

H. Havinga

F. Kuipers

D-J. Reijngoud

ADAPTED FROM J BIOL CHEM. 2008 12;283(37):25437-45

2Lxrα defi ciency hampers the hepatic adaptive response to fasting in mice

2 6 LXRα mediates the hepatic response to fasting

ABSTRACTBesides its well-established role in control of cellular cholesterol homeostasis, LXR has been impli-

cated in the regulation of hepatic gluconeogenesis. We investigated the role of the major hepatic

LXR isoform in hepatic glucose metabolism during the feeding-to-fasting transition in vivo. In addi-

tion, we explored hepatic glucose sensing by LXR upon carbohydrate refeeding.

Lxrα-/- mice and their wild-type littermates were subjected to a fasting-refeeding protocol and

hepatic carbohydrate fl uxes as well as whole-body insulin sensitivity were determined in vivo by

stable isotope procedures. Lxrα-/- mice showed an impaired response to fasting in terms of hepatic

glycogen depletion and TG accumulation. Hepatic G6P turnover was reduced in 9h-fasted Lxrα-/-

mice as compared to controls. Although hepatic gluconeogenic gene expression was increased

in 9h-fasted Lxrα-/- mice compared to wild-type controls, the actual gluconeogenic fl ux was not

aff ected by Lxrα defi ciency. Hepatic and peripheral insulin sensitivity were similar in Lxrα-/- and wild-

type mice. Compared to wild-type controls, the induction of hepatic lipogenic gene expression was

blunted in carbohydrate-refed Lxrα-/- mice, which was associated with lower plasma TG concentra-

tions. Yet, expression of ‘classic’ LXR target genes Abca1, Abcg5 and Abcg8 was not aff ected by Lxrα

defi ciency in carbohydrate-refed mice.

In summary, these studies identify LXRα as a physiologically relevant mediator of the hepatic re-

sponse to fasting. However, the data do not support a role for LXR in hepatic glucose sensing.

2 7Chapter 2

INTRODUCTIONLXR alpha and beta (LXRα/ß; NR1H3/NR1H2) are important players in the transcriptional control of

various metabolic pathways. LXRα is predominantly expressed in liver, intestine and adipose tissue,

but is also present in kidney, lung, and spleen. LXRβ is expressed in almost all tissues and organs

[54,55]. LXRs can be activated by oxidized cholesterol metabolites (oxysterols), which have been

identifi ed as their natural ligands. Hence, LXRs act as intracellular ‘cholesterol sensors’ [56]. LXRs in-

duce lipogenic gene expression upon activation, both directly [57] and indirectly via the transcrip-

tion factors SREBP-1c and ChREBP [26,57–59]. Both SREBP-1c and ChREBP control the conversion of

glucose into fatty acids. Thus, LXRs coordinate the interactions between sterol and fatty acid me-

tabolism, for instance to enable cholesterol ester formation during cellular cholesterol overload. In

the past years, several studies have been published that point toward a role of LXRs in the control of

glucose homeostasis. These studies showed that pharmacological LXR activation improves glycemic

control in diabetic rodent models by increasing peripheral glucose disposal [60,61] and/or inhibi-

tion of hepatic gluconeogenesis [61–64]. Mitro et al. recently reported that physiologically relevant

concentrations of either glucose or G6P are able to bind and activate LXR in HepG2 cells [53]. The

physiological relevance of this potential ´glucose sensing´ role of LXR has been debated [65–67] and

needs to be established.

In order to explore the physiological relevance of LXR in hepatic glucose metabolism we sub-

jected mice defi cient for Lxrα, the major hepatic isoform, to a fasting-refeeding protocol. Lxrα-/- mice

showed an impaired hepatic fasting response in terms of glycogen depletion and TG accumulation.

Although gluconeogenic gene expression was increased in 9-h fasted Lxrα-/- mice compared to wild-

type mice, stable isotope infusion revealed the actual gluconeogenic fl ux was not aff ected by Lxrα

defi ciency. G6P turnover was reduced in Lxrα-/- mice compared to wild-type mice. In carbohydrate-

refed Lxrα-/- mice, the hepatic lipogenic response was blunted while changes in the expression of the

LXR target genes Abca1, Abcg5 and Abcg8 were similar in wild-type and Lxrα-/- mice. Taken together,

these data imply an important role for LXRα in the control of hepatic glucose metabolism upon fast-

ing but they do not support the hypothesis that LXRα acts as a hepatic glucose sensor.

EXPERIMENTAL PROCEDURESAnimals and diets

F2 male Lxrα-/- mice and their wild-type littermates on a Sv129/OlaHsd C57Bl/6J mixed background

[68] were housed in a light- and temperature-controlled facility (lights on 7 AM-7 PM, 21 °C). They

were fed standard laboratory chow ad libitum (RMH-B, Abdiets, Woerden, The Netherlands) and had

free access to water. All experiments were approved by the Ethics Committee for Animal Experi-

ments of the University of Groningen.

Fasting and refeeding experiments

For fasting experiments we studied separate groups of mice. All mice were killed by cardiac punc-

ture under isofl urane anaesthesia at 8 AM, either without being fasted, after a 9-h fast, or after a 24-h

fast. For the refeeding experiments, mice were killed at 8 AM after a 24-h refeeding period with free

access to high carbohydrate chow (38.5% w/w sucrose, Abdiets) following a 24-h fasting period.


Plasma metabolite concentrations

Blood glucose concentrations were measured using a EuroFlash glucose meter (Lifescan Benelux,

Beerse, Belgium). Plasma insulin concentrations were determined using ELISA (Ultrasensitive Mouse

Insulin kit; Mercodia, Uppsala, Sweden). Plasma NEFA, β-hydroxybutyrate (β-HB), TG and cholesterol

concentrations were determined using commercially available kits (Roche Diagnostics, Mannheim,

Germany and Wako Chemicals, Neuss, Germany).

Hepatic metabolite content and gene expression levels

Livers were quickly removed, weighed, freeze-clamped and stored at -80 °C. A small piece of liver

was fi xed in 4% formalin in PBS for histological analysis. Blood was centrifuged (4000 g for 10 minutes

at 4 °C) and plasma was stored at -20 °C. Frozen liver was homogenized in ice-cold saline. Hepatic

TG concentrations were analyzed using a commercially available kit (Roche Diagnostics) after lipid

extraction according to Bligh and Dyer [69]. Hepatic G6P and glycogen content were determined as

previously described [70,71]. In addition, hepatic glycogen disposition was visualized by PAS staining

of 3 μm thick liver slices. RNA was extracted from frozen liver using TRI Reagent (Sigma, Zwijndrecht,

The Netherlands) and subsequently converted into cDNA by a reverse transcription procedure using

M-MLV and random primers according to the manufacturer’s protocol. For quantitative PCR (qPCR),

cDNA was amplifi ed using the appropriate primers and probes. Primer and probe sequences for

18S, ATP binding cassette a1/g5/g8 (Abca1/g5/g8, Chrebp, Fas, fructose-1,6-biphosphatase 1 (Fbp1),

G6ph, G6pt, peroxisome proliferator activated receptor gamma co-activator 1 alpha (Pgc-1α), Pepck,

Pdk4, Scd1 and Srebp-1c have been published (www.LabPediatricsRug.nl). The sequences of all other

primers and probes are given in Supplemental Table 1. All mRNA levels were normalized for 18S

expression.

In vivo fl ux measurements

Mice were equipped with a permanent catheter in the right atrium via the jugular vein [72] and were

allowed a recovery period of at least three days. After the recovery period, the mice were placed in

experimental cages and were fasted from 11 PM-8 AM with drinking water available. All infusion

experiments were performed in conscious, unrestrained mice. To determine hepatic carbohydrate

fl uxes, mice were infused with a solution containing [U-13C]glucose (7 μM), [2-13C]glycerol (82 μM),

[1-2H]galactose (17 μM) and paracetamol (1 mg/mL) during six hours at an infusion rate of 0.6 mL/h

as described previously [40,73]. Blood glucose concentrations were measured every 30 minutes.

Blood and urine spots were collected every 60 minutes on fi lter paper. In total, 80-90 μL of blood was

withdrawn per animal from the tail vein during these experiments.

Hyperinsulinemic euglycemic clamps were performed in a separate group of mice as described

earlier [60]. Mice were fasted from 11 PM-8 AM the next day with drinking water available. During six

hours, they were infused with two solutions. The fi rst solution contained bovine serum albumin (1%

w/v, Sigma), somatostatin (40 μg/mL, UCB, Breda, The Netherlands), insulin (110 mU/mL, Actrapid;

Novo Nordisk, Bagsvaerd, Denmark), glucose (1111 mM) and [U-13C]-glucose (33 mM, 99% 13C atom

%excess; Cambridge Isotope Laboratories, Andover, MA, USA) and was infused at a rate of 0.135

mL/h. The second solution consisted of glucose (1111 mM) containing [U-13C]-glucose (33 mM). The

infusion rate of this solution was variable to maintain euglycemia. Blood glucose concentrations

were measured every 15 minutes. Every 30 minutes, a bloodspot was collected. In total, 150-170 μL

of blood was withdrawn per animal from the tail vein during these experiments.

2 9Chapter 2

Analytical procedures for extraction of glucose from blood spots, derivatization of the extracted

compounds and GC-MS measurements of derivatives were performed according to van Dijk et al.

[39,40,73]. From this, hepatic carbohydrate fl uxes were calculated using mass isotopomer distri-

bution analysis (MIDA) as previously described [40,73]. Supplemental Figure 1 depicts the isotopic

model used. To balance input and output of hepatic G6P, minor adaptations were made to the pu-

blished equations [74]. The equations are given in Supplemental Table 2. Glucose production and

metabolic clearance rates during hyperinsulinemic euglycemic clamps were calculated according

to Grefhorst et al. [60].

Statistics

All data represent means ± SEM. Statistical analysis was performed using SPSS for Windows software

(SPSS 12.02, Chicago, IL, USA). Analysis of data obtained in Lxrα-/- versus wild-type mice was assessed

by Kruskal Wallis/Mann-Whitney U-test for plasma and liver parameters. In vivo fl ux data were ana-

lyzed by ANOVA for repeated measurements. Statistical signifi cance was reached at a p value below

0.05, except for the fasting-refeeding experiments, where this p value was adjusted for multiple

comparisons.


RESULTSWe compared the changes in metabolic parameters in fasted Lxrα-/- mice and wild-type littermate

controls. Upon fasting, blood glucose and plasma insulin concentrations decreased while plasma

NEFA and β-HB concentrations increased, without diff erences between Lxrα-/- and wild-type mice

(Table 1). Plasma TG concentrations increased upon fasting in both genotypes while plasma choles-

terol concentrations were not aff ected. Compared to wild-type mice, hepatic G6P content tended to

be higher in 9-h fasted Lxrα-/- mice (Figure 1A, +73%, p=0.26). Twenty-four hours of fasting decreased

hepatic G6P content in both phenotypes, but this drop was less pronounced in Lxrα-/- mice. Hepatic

glycogen content decreased upon fasting in both groups (Figure 1B). However, in wild-type mice

hepatic glycogen content already reached its lowest level after a 9-h fast, whereas in 9-h fasted Lxrα-/-

mice it was similar to what observed in the fed state. Histological analysis revealed that the glycogen

in the 9-h fasted Lxrα-/- mice was mainly located in the periportal zone (Figure 1C). After 24 hours of

fasting, hepatic glycogen stores were similarly depleted in both genotypes (Figure 1B). Hepatic TG

content increased upon fasting, but to a markedly less extent in Lxrα-/- mice compared to wild-type

controls (Figure 1D).

Table 1. Plasma parameters in Lxrα-/- mice and their wild-type littermates.

fed 9h-fasted 24h-fasted

wild-type Lxrα -/- wild-type Lxrα -/- wild-type Lxrα -/-

Blood glucose (mM) 8.8±0.3 9.0±0.7 5.2±0.3# 4.8±0.8# 3.5±0.4$ 3.6±0.5

Plasma insulin (ng/mL) 1.59±0.38 1.37±0.58 0.12±0.03# 0.13±0.1 0.06±0.01 0.06±0.02

Plasma NEFA (mM) 0.38±0.04 0.48±0.05 0.78±0.05# 0.68±0.04 0.84±0.03 0.78±0.04

Plasma ß-HB (mM) 0.18±0.04 0.23±0.09 1.57±0.31# 1.12±0.37 3.33±0.25$ 3.72±0.19$

Plasma TG (mM) 0.46±0.09 0.47±0.10 0.79±0.09 1.01±0.25 1.25±0.09$ 0.91±0.17

Plasma cholesterol (mM) 1.8±0.1 1.6±0.1 2.4±0.1# 1.9±0.1* 1.8±0.2 2.4±0.4

Blood glucose (mM) 8.8±0.3 9.0±0.7 5.2±0.3# 4.8±0.8# 3.5±0.4$ 3.6±0.5

Plasma insulin (ng/mL) 1.59±0.38 1.37±0.58 0.12±0.03# 0.13±0.1 0.06±0.01 0.06±0.02

Plasma NEFA (mM) 0.38±0.04 0.48±0.05 0.78±0.05# 0.68±0.04 0.84±0.03 0.78±0.04

Values represent means ± SEM for n=4-6; # p<0.05 9-h fasted vs. fed, $ p<0.05 24-h fasted vs. 9-h fasted, * p<0.05 Lxrα-/- vs. wild-type

(Mann-Whitney U-test, p value adjusted for multiple comparisons).

Gluconeogenic fl ux plays an essential role in glycogen accumulation [75] and hepatic gluconeoge-

nic gene expression, e.g. of Pepck and G6pase, has been shown to be decreased upon LXR activation

[61–63]. We therefore determined whether the increased hepatic glycogen content in the 9-h fasted

Lxrα-/- mice was paralleled by an increased expression of genes encoding enzymes involved in he-

patic gluconeogenesis. Compared to wild-type mice, hepatic expression of Pgc-1α, Pepck, Fbp1 and

G6ph (encoding G6P hydrolase, one component of the multi-protein complex G6Pase) were all in-

creased in 9-h fasted Lxrα-/- mice (Figure 2A). Expression of genes encoding other major enzymes in-

volved in hepatic carbohydrate metabolism (G6pt, Gk, Pk, Pdk4 and Gp, except for Gs Figure 2A/B) was

not aff ected by Lxrα defi ciency. Moreover, the lipogenic gene expression profi le was similar in 9-h

fasted wild-type and Lxrα-/- mice, except for a reduction of Acc2 and Scd1 expression (Figure 2C).

3 1Chapter 2

Impaired hepatic G6P metabolism in 9-h fasted Lxrα-/- mice is associated with decreased glucose

turnover and increased hepatic G6P content

A 9-h fast uncovered major diff erences in hepatic adaptive response between wild-type and Lxrα-/-

mice. To determine whether the increased gluconeogenic gene expression was a cause of the ob-

served diff erences in hepatic glycogen and G6P content between 9-h fasted wild-type and Lxrα-/-

mice, we determined glucose turnover, disposal and individual hepatic carbohydrate fl uxes using

stable isotope techniques [40,76]. During the infusion of the stable isotopes, blood glucose con-

centrations were lower in Lxrα-/- mice compared to wild-type littermates (Figure 3A). Steady state

isotope enrichment was reached from three hours of infusion onwards. Isotope dilution data during

this steady state situation are shown in Table 2. Glucose cycling and endogenous glucose produc-

tion were reduced in Lxrα-/- mice compared to their wild-type littermates (Figure 3B), resulting in a

decreased total glucose production. Metabolic glucose clearance was similar in both groups of mice

(Figure 3C).

fed 9h-fasted 24h-fasted0

200

400

600

800

1000 Lxrα -/-wild-type

Hepa

tic G

6P (n

mol

/g)

*$


50

100

150Lxrα -/-wild-type

Hepa

tic T

G (μ

mol

/g)

#

*

$

*

A


100

200

300

400


Hepa

tic g

lyco

gen

(μm

ol/g

)

B

wild-type Lxrα-/-

C D

Figure 1. Fasting response in Lxrα-/- mice and their wild-type littermates.

A, Hepatic G6P content. B, Hepatic glycogen content. C, Hepatic glycogen content and localization in 9-h fasted mice. P, periportal;

V, perivenous and D, Hepatic TG content.

Open bars, wild-type mice; fi lled bars, Lxrα-/- mice. Values represent means ± SEM for n=4-6; # p <0.05 9-h fasted vs. fed, $ p <0.05 24-h

fasted vs. 9-h fasted, * p <0.05 Lxrα-/- vs. wild-type (Mann-Whitney U-test, p value adjusted for multiple comparisons).


wild-type Lxrα -/-0

100

200

300

400endogenous glucose productionglucose cycling

Glu

cose

pro

duct

ion

(μm

ol/k

g/m

in)

**

0

10

20


Met

abol

ic c

lear

ance

rate

(mL/

kg/m

in)

0 1 2 3 4 5 60

5

10


Bloo

d gl

ucos

e (m

M)

*

Figure 3. Whole-body glucose fl uxes in 9-h fasted Lxrα-/-

mice and their wild-type littermates during steady state

infusion (t=180-360 min).

A, Blood glucose concentrations during isotope infusion.

Open dots, wild-type mice; fi lled dots, Lxrα-/- mice. B, To-

tal glucose production and contribution of endogenous

glucose production (dark grey bars) and glucose cycling

(light grey bars) and C, Metabolic glucose clearance rates.

Open bars, wild-type mice; fi lled bars, Lxrα-/- mice. Values re-

present means ± SEM for n=6; * p<0.05 Lxrα-/- vs. wild-type

(ANOVA).

BA

C

Pgc-1α

Pepck Fbp1G6ph

G6pt0

1

2

3

4

5

Lxrα -/-wild-type

Rela

tive

mRN

A ex

pres

sion *

* * *

Gk PkPdk4 Gs Gp

0

1

2

3

4

5

Lxrα -/-wild-type

Rela

tive

mRN

A ex

pres

sion

*

Srebp-1c

Acc1 Acc2 Fas Scd10

1

2

3

4

5

Lxrα -/-wild-type

Rela

tive

mRN

A ex

pres

sion

Figure 2. Hepatic gene expression levels in 9-h fasted Lxrα-/-

mice and their wild-type littermates.

A, Gluconeogenic gene expression. B, Glycolytic gene ex-

pression and C, Lipogenic gene expression.

Open bars, wild-type mice; fi lled bars, Lxrα-/- mice. Values re-

present means ± SEM for n=5; * p<0.05 Lxrα-/- vs. wild-type

(Mann-Whitney U-test).

BA

C

3 3Chapter 2

Table 2. Primary isotopic parameters during steady state infusion (t=180-360 min) in 9-h fasted Lxrα-/- mice and their wild-type litter-

mates.

wild-type Lxrα -/-

Isotope dilution

d(glc) 0.016±0.001 0.019±0.001*

d(UDPglc) 0.141±0.008 0.164±0.006*

Isotope exchange

c(glc) 0.27±0.02 0.19±0.01*

c(UDPglc) 0.15±0.03 0.11±0.01

MIDA analysis

f(glc) 0.66±0.03 0.71±0.02

f(UDPglc) 0.54±0.02 0.52±0.01

Values represent means ± SEM for n=6; * p<0.05 Lxrα-/- vs. wild-type (ANOVA).

For abbreviations see Supplemental Table 2.

Gluconeogenic fl ux, i.e., de novo synthesis of G6P was not aff ected by Lxrα defi ciency (Table 3). In ad-

dition, the compartmentation of newly synthesized G6P towards glucose (86±1% in both Lxrα-/- and

wild-type mice) and glycogen (14±1% in both Lxrα-/- and wild-type mice) was comparable in both

genotypes. However, glucose phosphorylation (glucokinase fl ux), dephosphorylation (glucose-6-

phosphatase fl ux), glycogen synthesis (glycogen synthase fl ux) and glycogen breakdown (glycogen

phosphorylase fl ux) were reduced in Lxrα-/- mice compared to wild-type mice (Table 3). G6P turnover

and glucose balance were reduced in Lxrα-/- mice compared to wild-type littermates, while glycogen

balance tended to be less negative in Lxrα-/- mice (Figure 4).

Table 3. Individual fl uxes comprising hepatic G6P metabolism during steady state infusion (t=180-360 min) in 9-h fasted Lxrα-/- mice and

their wild-type littermates.

wild-type Lxrα -/-

Gluconeogenic fl ux 109±6 98±4

Glucokinase fl ux 75±9 39±3*

Glucose-6-phosphatase fl ux 223±16 158±6*

Glycogen synthase fl ux 45±4 34±2*

Glycogen phosphorylase fl ux 71±7 48±4*

c(UDPglc) 0.15±0.03 0.11±0.01

Values represent means in μmol/kg/min ± SEM for n=6; * p<0.05 Lxrα-/- vs. wild-type (ANOVA).


0

5

10Lxrα -/-wild-type

Bloo

d gl

ucos

e (m

M)

A

0

50

100


Gluc

ose

prod

uctio

n ra

te

(μm

ol/k

g/m

in)

C

0

500


Gluc

ose

infu

sion

rate

(μm

ol/k

g/m

in)

B

0

50

100


Met

abol

ic c

lear

ance

rate

(mL/

kg/m

in)

D

Figure 5. Glucose metabolism under hyperinsulinemic euglycemic clamp conditions in 9-h fasted Lxrα-/- mice and their wild-type lit-

termates during steady state infusion (t=180-360 min).

A, Blood glucose concentrations. B, Glucose infusion rates required to maintain euglycemia. C, Endogenous glucose production rates

and D, Metabolic glucose clearance rates.

Open bars, wild-type mice; fi lled bars, Lxrα-/- mice. Values represent means ± SEM for n=5.

triose phosphate

glucose glucose-6-phosphate glycogen

Lxrα -/-

wild-type

0

100

200

Gluc

oneo

geni

c flu

x (μ

mol

/kg/

min

)

0

100

200

Gluc

ose

bala

nce

(μm

ol/k

g/m

in)

*

-25

0

Glyc

ogen

bal

ance

(μm

ol/k

g/m

in)

0

100

200

300

400

G6P

turn

over

(μm

ol/k

g/m

in)

*

Figure 4. Hepatic glucose balance, glycogen balance, G6P turnover and gluconeogenic fl ux in 9-h fasted Lxrα-/- mice and their wild-type

littermates during steady state infusion (t=180-360 min).

Open bars, wild-type mice; fi lled bars, Lxrα-/- mice. Values represent means ± SEM for n=6; * p<0.05 Lxrα-/- vs. wild-type (ANOVA).

3 5Chapter 2

Hepatic and peripheral insulin sensitivity are maintained in Lxrα-/- mice

Insulin is a major regulator of carbohydrate metabolism. Although plasma insulin concentrations

did not diff er between 9-h fasted wild-type and Lxrα-/- mice (Table 1), we questioned whether insulin

sensitivity of hepatic and peripheral glucose metabolism was altered in Lxrα-/- mice. We therefore

performed hyperinsulinemic euglycemic clamps in 9-h fasted conscious, unrestrained mice. Steady

state isotope enrichment and euglycemia (Figure 5A) were reached within three hours of infusion.

The glucose infusion rates to maintain euglycemic conditions (Figure 5B) did not diff er between the

two genotypes, indicative for unaff ected whole-body insulin-sensitivity in Lxrα-/- mice compared to

wild-type littermates. Hepatic insulin sensitivity was not aff ected in Lxrα-/- mice. Hyperinsulinemia

resulted in a 41% and 51% reduction of hepatic glucose production in Lxrα-/- and wild-type mice,

respectively (compare Figure 5C with Figure 3B). In addition, peripheral insulin sensitivity was not af-

fected by Lxrα defi ciency since the MCR was increased to 406% in Lxrα-/- mice and 378% in wild-type

littermates (compare Figure 5D with Figure 3C).

Carbohydrate refeeding aff ects hepatic lipogenesis and gene transcription independent of LXRα

We also determined whether there are indications for glucose-mediated LXR activation. Therefore,

plasma and liver metabolite concentrations were assessed in Lxrα-/- and wild-type mice that were

refed a carbohydrate rich diet following a 24-h fast (Table 4). Blood glucose and plasma insulin, NEFA

and β-HB concentrations were comparable in both groups of carbohydrate-refed mice. Plasma TG

concentrations were lower in carbohydrate-refed Lxrα-/- mice compared to wild-type mice, while

plasma cholesterol concentrations were similar. Hepatic G6P and glycogen content were increased

in carbohydrate-refed mice compared to mice that had been fasted for 24 hours (Figures 2A and 2B)

but no diff erences were observed between the two genotypes (Table 4). Hepatic TG content was

lower in carbohydrate-refed Lxrα-/- mice (p=0.052).

Table 4. Plasma and liver parameters upon refeeding in Lxrα-/- mice and their wild-type littermates.

wild-type Lxrα -/-

Blood glucose (mM) 9.5±0.5 9.8±0.6

Plasma insulin (ng/mL) 1.66±0.49 2.74±0.66

Plasma NEFA (mM) 0.34±0.01 0.30±0.02

Plasma ß-HB (mM) 0.11±0.01 0.12±0.01

Plasma TG (mM) 3.00±0.18 1.78±0.23*

Plasma cholesterol (mM) 3.7±0.2 3.3±0.1

Hepatic G6P (nmol/g) 347±23 359±41

Hepatic glycogen (μmol/g) 1121±81 1088±91

Hepatic TG (μmol/g) 18.1±1.5 13.0±1.6

Values represent means ± SEM for n=5-6, * p<0.05 Lxrα-/- vs. wild-type (Mann-Whitney U-test).


In both groups of mice, carbohydrate refeeding increased expression of Gk, Pk, and Gp, while Pdk4

expression was decreased. Chrebp and Gs expression were not aff ected by carbohydrate refeeding

(Figure 6A). Expression of Srebp-1c, Acc1, Fas and Scd1 was clearly induced in carbohydrate-refed

wild-type mice, but this response was less pronounced in Lxrα-/- mice. Acc2 expression was not af-

fected by carbohydrate refeeding (Figure 6B). In both wild-type and Lxrα-/- mice, expression of the

LXR target genes Abca1, Abcg5 and Abcg8 was not induced by carbohydrate-refeeding (Figure 6C).

B

Srebp-1c Acc1 Acc2 Fas Scd10

25

50

75

100wild-type 24hLxr α -/- 24hwild-type refedLxr α -/- refed

Rela

tive

mRN

A ex

pres

sion

&

&

& &

&&

&&

*

Chrebp Gk Pk

Pdk4 Gs Gp0

10

20

30

wild-type 24hLxr α -/- 24hwild-type refedLxr α -/- refed

Rela

tive

mRN

A ex

pres

sion

&&

&

&

&

&

&&

Abca1Abcg5

Abcg80

1

2

3

4

5wild-type 24hLxr α -/- 24hwild-type refedLxr α

Rela

tive

mRN

A ex

pres

sion -/- 24h

C

Figure 6. Hepatic gene expression levels upon fasting and

refeeding in Lxrα-/- mice and their wild-type littermates.

A, Glycolytic gene expression. B, Lipogenic gene expression.

C, Cholesterol transporter gene expression.

Light grey bars, 24-h fasted wild-type mice; dark grey bars,

24-h fasted Lxrα-/- mice; open bars, refed wild-type mice; fi l-

led bars, refed Lxrα-/- mice. Values represent means ± SEM for

n=5; & p<0.05 refed vs. 24-h fasted * p<0.05 Lxrα-/- vs. wild-

type (Mann-Whitney U-test, p value adjusted for multiple

comparisons).

A

3 7Chapter 2

DISCUSSIONLXRs act as cholesterol sensors that control transcription of genes involved in cellular cholesterol

and lipid homeostasis. Lipid and carbohydrate metabolism are tightly linked and strongly regulated

to ensure an adequate control of whole-body energy metabolism. LXR regulates transcription and

activity of the glucose-sensing lipogenic transcription-factor ChREBP [57], which strongly suggest a

physiological role of LXR in hepatic carbohydrate metabolism in the postprandial state. It is known

that LXR activation results in hepatic steatosis [51,58]. On the other hand, prolonged fasting is also

associated with hepatic lipid accumulation [9]. These lines of evidence prompted us to study the

role of hepatic LXR during fasting and refeeding. LXRα is considered to be the major isoform regu-

lating lipogenic gene expression in the liver. Therefore, we subjected Lxrα-/- mice [68] to fasting and

refeeding protocols and we applied sophisticated stable isotope techniques to quantify hepatic

carbohydrate fl uxes in vivo in these mice.

We are the fi rst to show that Lxrα plays an important role in the feeding-to-fasting transition. Lxrα

defi ciency results in an impaired fasting response, indicated by a delayed fasting-induced hepatic

glycogen depletion and increased hepatic G6P content in 9-h fasted Lxrα-/- mice compared to wild-

type littermates. Moreover, the Lxrα-/- mice accumulated less hepatic TG upon fasting. Expression of

gluconeogenic genes was increased in 9-h fasted Lxrα-/- mice compared to wild-type littermates.

This is in agreement with the decreased expression of Pgc-1α, G6pase and Pepck upon pharmaco-

logical LXR activation [61–63]. However, evaluation of hepatic carbohydrate fl uxes in 9-h fasted mice

revealed that the induction of gluconeogenic gene expression in Lxrα-/- mice was not paralleled by

an increased gluconeogenic fl ux. Thus, there is a discrepancy between gene expression levels and

gluconeogenic fl ux in vivo [60]. This indicates that other factors such as precursor availability [77,78]

and post-transcriptional modifi cation of enzymes are important determinants that control hepatic

carbohydrate fl uxes in vivo.

Glucose phosphorylation and dephosphorylation as well as glycogen synthesis and breakdown

were reduced in Lxrα-/- mice compared to wild-type littermates. Thus, instead of an altered de novo

synthesis of G6P the inter-conversions of G6P, glucose and glycogen were clearly aff ected in 9-h

fasted Lxrα-/- mice. The net eff ect of the lower glycogen synthesis (-24%) and breakdown (-32%)

fl uxes in Lxrα-/- mice was a less negative glycogen balance, supporting the delayed glycogen deple-

tion observed upon fasting in the Lxrα-defi cient mice. The remaining glycogen was located in the

periportal zone. It is known that upon fasting, glycogen is initially degraded to G6P in periportal

hepatocytes. In perivenous hepatocytes, glycogen is predominantly broken down into pyruvate and

hence released as lactate (reviewed in [79]). Thus in the livers of 9-h fasted Lxrα-/- mice, less glycogen

was broken down, contributing to the reduced G6P turnover observed in these mice. The changes in

G6P and glycogen metabolism were not secondary to changes in hepatic gluconeogenesis [75,80],

since neither the gluconeogenic fl ux nor the partitioning of newly synthesized G6P towards glucose

and glycogen was aff ected by Lxrα defi ciency. In addition, the net eff ect of the lower glucokinase

and glucose-6-phosphatase fl uxes was a reduction in endogenous glucose production and glucose

cycling.

Glycogen synthesis and breakdown are regulated by several factors including insulin. Although

insulin concentrations were comparable in 9-h fasted Lxrα-/- mice and their wild-type littermates,

hepatic insulin sensitivity could have been altered by Lxrα defi ciency, explaining the diff erences


observed in hepatic G6P and glycogen content as well as their inter-conversions. Hepatic and pe-

ripheral insulin sensitivity were determined in 9-h fasted Lxrα-/- mice and their wild-type littermates

using hyperinsulinemic euglycemic clamps. Insulin sensitivity of both hepatic glucose production

and peripheral glucose disposal was not aff ected by Lxrα defi ciency. Although LXR agonists have

been implicated as potential insulin sensitizers [61,62,64], our data do not support a direct role of

LXR as a potential mediator of hepatic and peripheral insulin sensitivity [60]. However, many of the

studies performed on the role of LXR are based on pharmacological activation. In the Lxrα-/- mice

there may be some adaptations that prevent the endogenous ligand from increasing, of there may

be additional systems that compensate for the Lxrα defi ciency. The reduced hepatic carbohydrate

fl uxes could also be a result from an altered reliance on glucose versus fatty acids and/or a dif-

ferential energy demand in the Lxrα-/- mice during the feeding-to-fasting transition. In addition to

the delay in glycogen depletion observed upon fasting in the Lxrα-/- mice, these mice accumulated

remarkably less TG. Gene expression analysis provided indications for an increase in hepatic fatty

acid oxidation in fasted Lxrα-/- mice, which could explain this remarkable reduction in hepatic TG

accumulation (data not shown). However, additional in vivo studies are required to determine the

physiological relevance of these observations.

Finally, we explored the role of LXRα in glucose-induced hepatic lipogenesis. Upon refeeding,

hepatic TG content was lower and plasma TG levels were reduced in Lxrα-/- mice compared to wild-

types. Quite strikingly, no diff erences in Chrebp expression were observed between Lxrα-/- and wild-

type mice. This is in contrast to observations by Cha and Repa (4) which suggested that CHREBP

is a downstream target of LXRα. However, carbohydrate refeeding resulted in a less pronounced

induction of Srebp-1c and other lipogenic gene expression in the Lxrα-/- mice compared to the wild-

types. Considering our observation that Chrebp and Pk expression were similar in carbohydrate-refed

Lxrα-/-and wild-type mice, we conclude that the blunted lipogenic response in carbohydrate-refed

Lxrα-/- mice resulted from the reduced SREBP-1c activity secondary to Lxrα defi ciency. Apparently,

the relationship between LXR, ChREBP and SREBP-1c on the one hand and hepatic TG metabolism

on the other hand requires further investigation.

Recent in vitro studies have shown that glucose is able to bind and activate hepatic LXR [53], sug-

gesting that LXR may act as a putative hepatic ‘glucose sensor’. However, the physiological relevance

of glucose sensing by LXR has been debated [65–67] and therefore required further investigation.

In the studies performed by Mitro et al. [53], the expression of the cholesterol transporters that are

direct LXR targets, e.g., Abca1 and Abcg1 only marginally increased upon carbohydrate-refeeding,

whereas lipogenic mRNA expression was clearly induced.

We confi rmed that the expression of the ‘classic’ LXR-target genes Abca1, Abcg5, and Abcg8 was

not aff ected by carbohydrate-refeeding in Lxrα-/- mice. Thus, the eff ect of carbohydrate refeeding on

hepatic lipogenic gene expression was diff erent from that on expression of the cholesterol trans-

porters Abca1, Abcg5, and Abcg8. Similar results have been obtained by Denechaud et al. [67], who

showed no induction of hepatic Abcg1 and Abca1 mRNA expression in carbohydrate-refed mice

while lipogenic gene expression was induced. Moreover, in contrast to the blunted induction of

lipogenic gene expression, Abcg1 and Abca1 expression was not aff ected in carbohydrate-refed

Lxrαβ-/- mice compared to wild-type controls [67]. Taken together, these and our data provide strong

evidence that carbohydrate refeeding does not induce hepatic gene expression via LXR, and there-

3 9Chapter 2

fore question the physiological relevance of glucose sensing by hepatic LXR in vivo.

In summary, our data identify LXRα as an important player in control of metabolic adaptation dur-

ing the feeding-to-fasting transition, but question the physiological relevance of glucose sensing by

hepatic LXR. In addition to its regulatory role in cholesterol, lipid and glucose metabolism to ensure

energy storage in the postprandial state, LXRα seems to facilitate the release of stored energy upon

fasting. Under these conditions, LXRα not only mediates TG accumulation, but also controls hepatic

G6P and glycogen deposition as it determines the partitioning and turnover of these energy-bear-

ing molecules, possibly to fulfi ll the liver’s demand of these metabolites.

ACKNOWLEDGEMENTSThe authors thank Juul F.W. Baller for excellent technical assistance.

3 An increased fl ux through the glucose-6-phosphate pool in enterocytes

delays glucose absorption in Fxr -/- mice

T.H. van Dijk

A. Grefhorst

M.H. Oosterveer

V.W. Bloks

B. Staels

D-J. Reijngoud

F. Kuipers

ADAPTED FROMJ BIOL CHEM. 2009 17;284(16):10315-23

4 2 Delayed intestinal glucose absorption in Fxr -/- mice

ABSTRACTFXR is involved in regulation of bile acid and lipid metabolism. Recently, a role for FXR in control of

glucose metabolism became evident. Because FXR is expressed along the length of the small intes-

tine, we evaluated the potential role of FXR in glucose absorption and processing. During i.v. infusion

of a trace amount of D-[6,6-2H2]-glucose, a D-[U-13C]-glucose-enriched oral glucose bolus was given

and glucose kinetics were determined in wild-type and Fxr -/- mice. Compared to wild-type mice,

Fxr -/- mice showed a delayed plasma appearance of orally administered glucose. Multicompartmen-

tal kinetic modelling revealed that this delay was caused by an increased fl ux through the G6P pool

in enterocytes.

Thus, our results show involvement of FXR in intestinal glucose absorption, representing a novel

physiological function for this nuclear receptor.

4 3Chapter 3

INTRODUCTIONBile acid-activated FXR (NR1H4) is a member of the nuclear receptor superfamily that is expressed in

liver, adrenals, kidney, small intestine, and colon [81]. Through FXR activation in the liver, bile acids in-

duce transcription of the atypical nuclear receptor short heterodimer partner (SHP; NR0B2), which, in

turn, represses transcription of the Cyp7a1 gene, encoding the rate-controlling enzyme in bile acid

synthesis [82]. FXR also suppresses transcription of the gene encoding the hepatobiliary bile acid

uptake transporter NTCP (Slc10a1) and induces transcription of genes encoding canalicular bile acid

transporters such as the bile salt export pump (ABCB11) and MRP2 (ABCC2) (see reviews [81,83]). In

the intestine of mice, FXR stimulates transcription of the gene encoding the fi broblast growth factor

15 (FGF15) [84]. FGF15 reduces hepatic bile acid synthesis by repressing Cyp7a1 transcription in the

liver. Apart from its clearly established eff ects on bile acid synthesis and transport, FXR is involved

in the control of lipid and lipoprotein homeostasis. Fxr -/- mice have elevated plasma TG and choles-

terol levels [85] and FXR activation decreases plasma TG levels in mice [86]. FXR negatively controls

apoA-I [87] as well as apoCIII expression [86], which contributes to FXR-mediated control of plasma

lipid levels.

Recently, a link between FXR and glucose homeostasis has become evident. It was shown that

glucose induces hepatic Fxr expression in rodent liver, probably via intermediates of the PPP [88].

Recent publications indicate that FXR plays a role in the regulation of the transcription of various

hepatic carbohydrate metabolism-related genes. Activated FXR represses the transcription of glu-

coneogenic genes, e.g., Pepck, fructose-1,6-biphosphatase-1 (Fbp1), and G6Ph in vitro [89]. In vivo ex-

periments showed that Fxr -/- mice have a reduced peripheral insulin sensitivity [45,90] and a reduced

hepatic glucose production rate [44].

Intestinal glucose transport is an important determinant of blood glucose levels. After uptake of

glucose by the enterocyte, glucose can take either a direct or an indirect pathway through the cell.

In the indirect pathway, glucose is phosphorylated by HK1 or HK2 into G6P. The catalytic subunit of

glucose-6-phosphatase (G6PH) dephosphorylates G6P and makes glucose available for transport

across the basolateral membrane into the portal vein [91,92]. Recent studies revealing the localiza-

tion of the FXR protein in the absorptive epithelium of the small intestine [93] lead us to determine

the physiological impact of intestinal FXR on glucose homeostasis and intestinal glucose absorp-

tion. For this, oral D-[U-13C]-glucose tolerance tests (OGTT) were performed and the absorption of

glucose from the intestine was calculated in this non-steady-state situation.


EXPERIMENTAL PROCEDURESAnimals

The Fxr -/- mice were generated by Deltagen, Inc. (Redwood City, CA) [94]. Male 18-21 week old

Fxr -/- mice and wild-type littermates were housed in a light- and temperature controlled facility. The

animals were fed a commercially available lab chow (RMH-B, Hope Farms BV, Woerden, The Neth-

erlands). All experiments were approved by the Ethics Committee for Animal Experiments of the

University of Groningen.

Oral glucose tolerance test

After a 9-h fast (11 PM-8 AM), mice were given an oral glucose bolus of 5.6±0.2 mmol/kg in 0.2

mL water under light isofl urane anaesthesia. Blood glucose concentrations were determined in

2 μL blood drawn from the tail with a handheld Lifescan EuroFlash glucose meter (Lifescan Benelux,

Beerse, Belgium) at the indicated time points. A similar experiment was conducted to determine

the eff ect of the OGTT on plasma insulin concentrations. After a 9-h fast (11.00 pm to 8.00 am), mice

were given an oral glucose bolus of 5.6 ± 0.2 mmol/kg in 0.2 mL water under light isofl urane anaes-

thesia. About 40 μL of blood was collected by orbital bleeding shortly before the oral glucose bolus.

In addition, the same amounts of blood were drawn from the tail at the indicated time points after

the oral glucose bolus. Plasma insulin levels were measured using a commercially available ELISA kit

(Mercodia ultrasensitive mouse insulin Elisa, Orange Medical, Tilburg, the Netherlands).

Oral D-[U-13C]-glucose tolerance test combined with continuous D-[6,6-2H2]-glucose infusion

Mice were equipped with a permanent catheter in the right atrium via the jugular vein [72]. The

entrance of the catheter was attached to the skull with acrylic glue and the mice were allowed

to recover for a period of at least fi ve days. After a 9-h fast (11 PM-8 AM) in which they were kept

in metabolic cages, mice received a continuous infusion of 4.4 μmol/kg/min D-[6,6-2H2]-glucose

(Cambridge Isotope Laboratories, Andover, MA) for 5 hours to determine the total rate of glucose

appearance. To determine the appearance of intestine-derived glucose, an oral glucose bolus of

11.2±0.2 mmol/kg from which 30% was D-[U-13C]-glucose (Cambridge Isotope Laboratories) in 0.2

mL water was given under light isofl urane anaesthesia at 2 hours after start of the infusion. At in-

dicated time points, blood glucose levels were determined in 2 μL blood with a handheld Lifescan

EuroFlash glucose meter and 10 μL blood spots were taken from the tail on fi lter paper for analysis

of isotopic enrichments [40]. At the end of the experiment, the mice were killed by cardiac puncture

under isofl urane anaesthesia. Blood and livers were collected for further analysis.

Short term oral glucose tolerance test

After a 9-h fast (11 PM-8 AM), mice were either not treated or were given an oral glucose bolus of

11.2±0.2 mmol/kg in 0.2 mL water under light isofl urane anaesthesia After 30 minutes, mice were

killed by cardiac puncture under isofl urane anaesthesia. Blood and livers were collected for further

analysis. Small intestines were removed and rinsed with 10 mL saline and divided into three equal

sections. Samples were taken from the very fi rst part of the intestine and from the middle of each

intestinal section for measurements of mRNA expression level and metabolite concentrations.

GC-MS measurements

The fractional isotopomer distribution measured by GC-MS (m0-m

6) was corrected for the fractional

distribution due to natural abundance of 13C by multiple linear regression as described by Lee et al.

[39] to obtain the excess fractional distribution of mass isotopomers (M0-M

6) due to dilution of infu-

4 5Chapter 3

sed labeled compounds, i.e., D-[U-13C]-glucose and D-[6,6-2H2]-glucose. This distribution was used in

the calculations of blood glucose kinetics.

Single-pool, fi rst-order kinetic model

The excess fractional distribution of mass isotopomers was used to calculate the fi rst order absorp-

tion process in an one-compartment model [95-97] using SAAM-II software (version 1.2.1, SAAM

Institute, University of Washington, Seattle, WA). The formulas used to calculate the concentration vs.

time curves and the kinetic parameters are given in Supplemental Table 3.

Calculations of the glucose appearance rates

The total rate of glucose appearance (RaT) in blood was calculated applying a one-compartment

model according to Steele [98]. At time point t2 RaT

t2 = ((infusion(glc;M

2) x M

2(glc)

infuse) –Vs x ([glc]

2

– [glc]1)/2 x (M

2(glc)

blood,2 – M

2(glc)

blood,1)/(t

2 – t

1)) / M

2(glc)

blood,2. In this equation infusion(glc;M

2) is

the infusion rate of the D-[6,6-2H2]-glucose; M

2(glc)

infuse is the excess mole fraction of infused D-[6,6-

2H2]-glucose; Vs is the glucose distribution volume calculated from a pool fraction of 0.48 with a

total glucose volume of 0.222 mL/kg resulting a Vs of 0.48 x 0.222 L/kg (data generated with the

single-pool, fi rst-order kinetic model, see Supplemental Table 3); [glc]2 and [glc]

1 are blood glucose

concentrations at time points t2 and t

1, respectively; M

2(glc)

blood,2 and M

2(glc)

blood,1 are the the excess

mole fraction of blood D-[6,6-2H2]-glucose at time points t

2 and t

1, respectively. The total glucose vol-

ume was calculated in the single-pool, fi rst-order kinetic model. The area under the curve (AUC) was

estimated using the Trapezoidal Rule. The rate of appearance of exogenous glucose (RaE) was calcu-

lated according to Tissot et al. [99]. At time point t2 RaE

t2 = ((RaT

t2 x (M

6(glc)

blood,2 + M

6(glc)

blood,1)/2) +

(Vs x ([glc]1+[glc]

2)/2 x (M

6(glc)

blood,2 – M

6(glc)

blood,1)/(t

2-t

1)))/M

6(glc)

ingested. In this equation M

6(glc)

blood,2

and M6(glc)

blood,1 are the excess mole fractions of blood D-[U-13C]-glucose at time points t

2 and t

1,

respectively; M6(glc)

ingested is the excess mole fraction of ingested D-[U-13C]-glucose. By substracting

this value from the total rate of appearance, the endogenous glucose production (EGP) was calcu-

lated [96]: EGP = RaT – RaE.

Metabolite contents and gene expression levels

Hepatic and intestinal G6P and glycogen content were determined as described previously [70,71].

RNA was isolated from liver and intestinal tissue using the Trizol method (Invitrogen, Paisley, UK).

Using random primers, RNA was converted to cDNA with M-Mulv-RT (Roche Diagnostics, Mannhein,

Germany) according to the manufacturer’s protocol. The cDNA levels of the genes of interest were

measured by RT-PCR using the ABI Prism 7700 Sequence Detection System (Applied Biosystems,

Foster City, CA). An amount of cDNA equivalent to 20 ng of total RNA was amplifi ed using the qPCR

core kit (Eurogentec, Seraing, Belgium) according to the manufacturers protocol with the appro-

priate forward and reverse primers (Invitrogen) and a template-specifi c 3’-TAMRA, 5’-FAM-labeled

Double Dye Oligonucleotide probe (Eurogentec). Calibration curves were run on serial dilutions of

pooled cDNA solutions as used in the assay. The data were processed using ABI Sequence Detector

v.1.6.3. in the linear part of the calibration curves. PCR results were normalised by β-actin RNA levels.

Primer and probe sequences Hk1, Hk2, G6ph, G6pt and Glut2 have been published (www.labpediat-

ricsrug.nl). The sequences for all other primers and probes are given in Supplemental Table 1.


Statistics

All values represent means ± SEM. Statistical diff erences were determined using one-way ANOVA or

Kruskal Wallis/Mann-Whitney U-test (metabolite concentrations and gene expression data). Statisti-

cal signifi cance was reached at a p value below 0.05.

RESULTSFxr -/- mice show an altered plasma glucose response during an OGTT

Although fed male Fxr -/- mice and wild-type littermates had comparable body weights, there was a

small but statistically signifi cant diff erence in weight loss upon 9 hour fasting between both groups

(Table 1). Blood glucose concentrations were signifi cantly lower in Fxr -/- than in wild-type mice,

before and after the 9-h fast. Upon the OGTT, the Fxr -/- mice had a reduced and delayed increase

of blood glucose concentrations compared to their wild-type controls with constituently higher

plasma insulin concentrations (Figure 1). This diff erence in blood glucose concentration between

the genotypes might be due to the fact that Fxr -/- mice have 1) an enhanced glucose disposal rate;

2) a reduced and/or delayed intestinal glucose absorption rate; and/or 3) a stronger reduction of the

endogenous glucose production rate upon OGTT. We therefore investigated blood glucose kinetics

in more detail in an experiment in which 30% of the oral glucose bolus was substituted by D-[U-13-

C]-glucose. This experiment was performed in combination with a continuous infusion of a trace

amount of D-[6,6-2H2]-glucose.

Before start of the D-[U-13C]-glucose-containing OGTT, the Fxr -/- mice had slightly lower blood

glucose concentrations compared to wild-type mice (Figure 2A). Within 15 minutes after oral glu-

cose administration, blood glucose concentrations rose to maximal levels of 19.0±1.3 mM in wild-

type and to 14.5±1.1 mM in Fxr -/- mice. These levels returned to pre-OGTT concentrations after 90

minutes. Blood D-[U-13C]-glucose vs. time curves were diff erent between both groups (Figure 2B).

Compared to wild-type mice, D-[U-13C]-glucose concentrations were lower in Fxr -/- mice during the

fi rst 45 minutes but higher during the last part of the experiment. Applying the formulas for single-

pool, fi rst-order kinetics (Supplemental Table 3), curves were fi tted for each individual mouse. Figure

2B shows the averages of data points and estimated curves whereas the estimated and derived

parameters are presented in Table 2.

Extrapolated blood D-[U-13C]-glucose concentrations at t=0 (C(0)el) and (C(0)ab), were signifi cantly

lower in Fxr -/- mice compared to wild-types. A signifi cantly lower elimination rate constant (kel) was

also evident without diff erences in absorption rate constant (kab). This observation clearly falsifi es

the fi rst hypothesis to explain the perturbed blood-glucose curve in Fxr -/- mice upon the OGTT.

Surprisingly, signifi cantly higher values were calculated for the apparent bioavailability of the oral

glucose dose (F) in the Fxr -/- mice. This higher F can be explained by a more gradual introduction of

D-[U-13C]-glucose into the blood compartment, supporting our second hypothesis: Fxr -/- mice have

a reduced and/or delayed intestinal glucose absorption.

4 7Chapter 3

Table 1. Body weight and blood glucose concentrations before and after 9 hours of fasting in wild-type and Fxr -/- mice.

wild-type Fxr -/-

Fed body weight (g) 32.8±1.0 33.9±1.0

Fasted body weight (g) 29.9±0.9 30.1±0.9

Weight loss (g) 2.9±0.1 3.8±0.1*

Weight loss (%) 8.8±0.3 11.2±0.9*

Fed blood glucose (mM) 9.5±0.6 6.3±0.3*

Fasted blood glucose (mM) 6.4±0.6 5.1±0.5*

Values represent means ± SEM for n = 12 (wild-types) and n = 10 (Fxr -/-); *, p< 0.05 (Mann-Whitney U-test).

0

250

500

750

AU

C

(m

M/m

in)

0-1

20

**

Fxr -/-Fxr +/+

Time (hours)

Bloo

d gl

ucos

e (m

M)

0

5

10

15

0 1 2 3

***

Fxr +/+

Fxr -/-

0

2

4

6

0 1 2

*

Fxr +/+

Fxr -/-

*

Time (hours)

Plas

ma

insu

lin (m

U/L

)

A B

C

Figure 1. Blood glucose concentrations during an OGTT.

A, Blood glucose concentrations during the OGTT. B, The

area under curve of the excess of blood glucose concentra-

tion (baseline is timepoint 0 and timepoints 120 – 180 minu-

tes) and C, Plasma insulin concentrations during the OGTT.

Values represent means ± SEM for n = 5; *, p<0.05; **, p<0.01

(ANOVA).

0

1

2

3

**

*** *

Bloo

d D

-[U-

13C]

-glu

cose

(mM

)

Time (hours)

Fxr +/+

Fxr -/-

-1 0 1 2 3

13C]-glucose bolusD -[U -

0

10

20

-1 0 1 2 3

Bloo

d gl

ucos

e (m

M)

Fxr +/+

Fxr -/-

Time (hours)

D -[U -13C]-glucose bolusD -[U -

A B

Figure 2. Blood glucose kinetics before and during OGTT using the fi rst-order, one-compartment model.

A, Blood glucose concentrations before and during the OGTT and B, Calculated blood D-[U-13C]-glucose concentrations from the

fractional contribution of D-[U-13C]-glucose with the estimated curve using SAAM II software. Values represent means ± SEM for n =

5; *, p<0.05; **, p<0.01 (ANOVA).


Table 2. Glucose kinetics during an OGTT in wild-type and Fxr -/- mice.

wild-type Fxr -/-

DL (mmol/kg) 3.47±0.09 3.34±0.10

C(0)el (mM) 9.26±0.44 3.91±0.123**

kel (min-1) 0.0331±0.0005 0.0201±0.0008**

C(0)ab (mM) 13.16±0.41 4.74±0.24**

kab (min-1) 0.091±0.007 0.100±0.007

tlag

(min) 6.0±0.4 2.4±0.6**

Clag

(mM) 7.60±0.33 3.72±0.13**

tmax

(min) 17.4±0.8 20.1±1.0*

Cmax

(mM) 2.72±0.24 1.99± 0.07*

t(½)el (min) 20.9±0.2 34.5±1.2**

MRT (min) 41.1±1.0 59.7±1.9**

F 0.48±0.04 0.76±0.02**

DL, oral dose administrated D-[U-13C]-glucose; C(0)

e, initial D-[U-13C]-glucose concentration by extrapolation elimination period;

C(0)a, initial D-[U-13C]-glucose concentration by extrapolation absorption period; kel, elimination rate constant; kab, absorption rate

constant; tlag

, time between administration and appearance of D-[U-13C]-glucose in sampled compartment; Clag

, concentration at lag

time calculated from elimination or absorption curve; tmax

, time of maximal D-[U-13C]-glucose concentration; Cmax

, D-[U-13C]-glucose

concentration at tmax

. t(½)el, half-life of blood glucose; MRT, mean residence time of glucose in sampled compartment. F, fractional

contribution of administered D-[U-13C]-glucose to the sampled compartment; VD, appearant volume of distribution of administered

D-[U-13C]-glucose.

Values represent means ± SEM for n=5; *, p<0.05; **, p<0.01 (ANOVA).

Fxr -/- mice show delayed intestinal glucose absorption after an OGTT

To elucidate why blood glucose levels were markedly less increased after an OGTT in Fxr -/- mice, we

used non-steady state equations according to Steele [98].

The continuous infusion of D-[6,6-2H2]-glucose enabled us to calculate blood glucose turnover

rates during the experiment. Fxr -/- mice had a signifi cant reduced baseline glucose turnover com-

pared to wild-type mice (102.4±6.8 vs. 128.3±8.3 μmol/kg/min, p<0.05) (Figure 3A). After correction

for baseline values (the average from time points -30 - 0 and 120 – 180 minutes), it was clear that the

OGTT-mediated increase of glucose appearance rate was also decreased in Fxr -/- mice (185.6±20.1 vs.

110.8±16.6 μmol/kg/min, Fxr -/- vs. wild-type, p<0.05) (Figure 3B). In addition, the decline of glucose

rate of appearance to baseline values was slower in Fxr -/- mice.

Because the oral glucose bolus contained a diff erent stable isotope of glucose than the infusate,

we were able to calculate the appearance rate of glucose derived from the intestine. Wild-type mice

showed a steep, isolated peak in intestinal glucose absorption whereas Fxr -/- mice showed a blunted

absorption rate with a much slower decrease to baseline (Figure 3C). In Fxr -/- mice, the reduced

recovery of intestine-derived glucose in the fi rst 45 minutes after the oral glucose bolus was fully

compensated for in the period thereafter (Figure 3D). Altogether, this resulted in similar recoveries at

the end of the test. These data show that Fxr -/- mice had delayed but not reduced intestinal glucose

absorption after OGTT.

4 9Chapter 3

The diff erence between the total rate of glucose appearance and the appearance of intestine-de-

rived glucose gives the EGP. Baseline EGP was signifi cantly lower in Fxr -/- mice than in wild-type mice

(Figure 3E). The EGP was signifi cantly reduced in both groups upon administration of the glucose

bolus but the reduction was more pronounced in wild-type mice. The relative reduction was 50±3%

vs. 30±4% in wild-type and Fxr -/- mice, respectively. This latter observation of reduced reduction in

EGP is inconsistent with our third hypothesis that EGP could be more reduced upon the OGTT in

Fxr -/- than in wild-type mice.

Reduced intestinal glucose absorbance in Fxr -/- mice is likely the result of increased glucose

phosphorylation in proximal enterocytes

From the three proposed mechanisms that might explain the delayed increase of blood glucose in

Fxr -/- mice upon the OGTT, only a delayed intestinal glucose absorption rate seems to be valid. We

next tried to unravel the cause of this hampered glucose absorption and tested whether enterocytic

glucose handling might be disturbed in Fxr -/- mice. Therefore, glucose, glycogen, and G6P contents

in the liver and small intestine were measured after the absorptive phase of the OGTT. Animals were

sacrifi ced when the blood glucose values reached their peak (30 minutes after glucose administra-

tion) and at the end of the test (180 minutes after glucose administration). At both time points

hardly any administered glucose was found in the intestinal lumen of both groups, indicating a

complete uptake in all mice (Figure 4A). Fxr -/- mice had lower hepatic glycogen and G6P contents

than wild-type mice 30 minutes after the oral glucose bolus (Figures 4B and 4C). In 9-hour fasted

mice, G6P concentrations in the proximal section of the small intestine were lower in the Fxr -/- mice

than in wild-type mice (Figure 4D). Enterocyte G6P concentrations were not aff ected by the OGTT

in wild-type mice. In Fxr -/- mice, however, they signifi cantly increased to values comparable to that

of wild-type mice.

Next, we compared the expression of genes encoding proteins involved in intestinal glucose ab-

sorption and metabolism in sequential parts of the small intestine. Expression patterns of the gene

encoding the brush border located sodium-dependent glucose/galactose transporter 1 (SGLT1;

Slc5A1) and the basolaterally located GLUT2 (Slc2A2) was similar in the Fxr -/- and wild-type mice (Fig-

ures 5A and 5B). Compared to wild-type mice, the expression of the genes encoding the glucose-

phosphorylating enzymes HK1 and HK2 was signifi cantly increased in the proximal part of the small

intestine of Fxr -/- mice (Figures 5C and 5D). Expression of G6ph and the gene encoding G6PT did

not diff er between both genotypes, although both tended to be lower in Fxr -/- mice compared to

wild-types (Figures 5E and 5F). Combined, these data suggest that delayed glucose passage through

proximal enterocytes of Fxr -/- mice is likely the result of an increased glucose phosphorylation.

Diverted glucose fl ux through G6P pool in enterocytes of Fxr -/- mice

The metabolic and gene expression data are indicative for an enhanced fl ux of glucose through

G6P in the enterocyte of Fxr -/- mice compared to wild-type mice. We therefore considered it feasable

to address the process of intestinal glucose absorption using a compartmental model (build using

SAAM II software) comprising the direct (without intracellular metabolism) and indirect pathways

(comprising the HK and G6Pase reactions) (Figure 6A) as described by Stumpel et al. [91]. The fi t

between the simulated appearance of D-[U-13C]-glucose in the circulation (Figure 6B) was obtained

when the initial direct fl ux was calculated to be equal to 187 μmol/kg/min in the wild-type mice

and lower (134 μmol/kg/min) in the Fxr -/- mice (Figure 6C). In contrast to the decreased direct fl ux in


Fxr -/- mice, the values for the fl ux through both the HK and G6Pase was increased in the Fxr -/- mice

(Figures 6D and 6E). The sum of the direct and G6Pase fl uxes representing the total fl ux resulted

in a glucose fl ux that is clearly reduced in Fxr -/- mice (Figure 6F), especially in the fi rst 30 minutes

after glucose administration. The compartmental model shows that the D-[U-13C]-glucose disposal

rate was initially lower in Fxr -/- mice, but was compensated at the end of the experiment (Figure

6G). It can be concluded from this simulation study that in enterocytes of wild-type mice glucose

is absorbed preferentially by the direct pathway. In enterocytes of Fxr -/- mice, the indirect pathway

becomes equally important.

Figure 3. Glucose appearance rates before and during

OGTT.

A, Total rate of glucose appearance in blood before and

during OGTT. B, Total rate of glucose appearance in blood

before and during OGTT after correction for baseline glu-

cose appearance (baseline is time points -1 – 0 hour and

2 – 3 hour). C, Rate of appearance of orally administrated

glucose. D, Fractional recovery of orally administrated glu-

cose over time with fractional recovery of orally adminis-

tered glucose during the fi rst 45 minutes and during the

45 – 180 minutes after oral glucose bolus (inset) and E, En-

dogenous glucose production rate with baseline values

(inset). Values represent means ± SEM for n = 5; *, p<0.05;

**, p<0.01 (ANOVA).

100

200

300

-1 0 1 2 30

Time (hours)

Tota

l glu

cose

app

eara

nce

rate

(μm

ol/k

g/m

in)

FxrFxrFxr +/+

Fxr -/-**

D -[U -13 C]-glucose bolusD -[U -

A

0

50

100

-1 0 1 2 3

Reco

very

of a

dmin

iste

red

D-[

U-13

C]-g

luco

se (%

)

FxrFxrFxr +/+

Fxr -/-

Time (hours)

D -[U -13C] - glucose bolusD -[U -

*

*0

2550

0’-45’ 45’-180’Reco

very

(%)

* *

0

50

100

150

-1 0 1 2 3

Endo

geno

usgl

ucos

e ap

pear

ance

rate

( μ

mol

/kg/

min

)

FxrFxrFxr +/+

Fxr -/-

Time (hours)


0

50

100

Base

line E

GP(μ

mol/

kg/m

in)

Fxr +/+ Fxr -/-

*E

-

0

50

100

150

-1 0 1 2 350

-1Co

rrec

ted

tota

lgl

ucos

e ap

pear

ance

rate

(μ

mol

/kg/

min

)

FxrFxrFxr +/+

Fxr -/-**

*

Time (hours)


B

0

100

200

0 1 2 3-1

App

eara

nce

rate

of

oral

ly a

dmin

iste

red

gluc

ose

(μm

ol/k

g/m

in)

FxrFxrFxr +/+

Fxr -/-

*

*

**

**

Time (hours)

D -[U -13 C]-glucose bolusD -[U -

C D

5 1Chapter 3

0.0

0.5

1.0

1.5

30 minutes 180 minutes

$$

Fxr

Fxr

Fxr

Fxr

Fxr +/+

Fxr -/-Fr

actio

nal r

ecov

ery

of

inge

sted

glu

cose

(%)

0

2

4

6

Fxr +/+ Fxr -/-

Live

r gly

coge

n

(μm

ol/g

pro

tein

)

0

2

4

6

**

Fxr +/+ Fxr -/-

Live

r G6P

(μm

ol/g

pro

tein

)

0.25

0.50

*

$

30 minutes

Fxr

Fxr

0.000 minutes

Fxr +/+

Fxr -/-

Prox

imal

inte

stin

al G

6P

(μm

ol/g

pro

tein

)

A

C D

Figure 4. Contents of metabolites in liver and intestine.

A, Fractional recovery of ingested glucose in the lumen of the small intestine 30 and 180 minutes after the oral glucose bolus. B, Hepatic

G6P concentration 30 minutes after the oral glucose bolus. C, Hepatic glycogen concentration 30 minutes after the oral glucose bolus

and D, Proximal intestinal G6P concentration 30 minutes after the oral glucose bolus compared to 9-h fasted mice. Values represent

means ± SEM for n = 5; *, p<0.05 between genotypes; **, p<0.01 between genotypes; $, p<0.05 between timepoints (Mann-Whitney

U-test).

B


Figure 5. Gene expression profi les of the small intestine.

Results are normalized to β-actin and to the most proximal part of the wild-type mice.

Values represent means ± SEM for n = 5; *, p<0.05; **, p<0.01 (Mann-Whitney U-test).

Sglt14

3

2

1

0Rela

tive

mRN

A ex

pres

sion

Proximal Distal

Fxr +/+

Fxr -/-

A Glut24

3

2

1

0 Rel

ativ

e m

RNA

expr

essi

on

Proximal Distal

B

*

Hk24

3

2

1

0Rela

tive

mRN

A ex

pres

sion

Proximal Distal

D* Hk14

3

2

1

0 Rel

ativ

e m

RNA

expr

essi

on

Proximal Distal

C

G6ph4

3

2

1

0

Rela

tive

mRN

A ex

pres

sion

Proximal Distal

E 4 G6pt

3

2

1

0Rela

tive

mRN

A ex

pres

sion

Proximal Distal

F

5 3Chapter 3

Figure 6. Estimated glucose fl uxes during OGTT using

compartmental modelling.

From the proposed model of glucose metabolism in ente-

rocyte (see references 91 and 92), a compartmental model

was made that was used in SAAM II software to calculate

fl uxes through the compartmental model. A. The used

compartimental model. The oral bolus was administrated

in the intestinal lumen of the mouse. After transport over

the brush border membrane the glucose can either leave

the enterocyte directly or it is phosphorylated by HK to

G6P. G6P, in turn, can be hydrolyzed in the ER to glucose

by G6Pase. Both the direct fl ux and the fl ux via G6P end

in the blood compartment where samples are taken. The

amount of glucose ingested has be corrected to get the

amount that enters the sampled pool (F). The volume

of distribution (VD) and the lag time (t

lag) also have to be

known. These three parameters are introduced in the “de-

lay compartment”. The used values for these parameters

were: F = 0.48; VD = 0.222 L/kg; t

lag = 6.0 min. B. Calcula-

ted concentrations of D-[U-13C]-glucose in the sampled

pool with the estimated curve. C. Direct fl ux from lumen

to blood compartment. D. Hexokinase fl ux. E. Glucose-6-

phosphatase fl ux. F. Flux to the sampled pool after correc-

tion for F, VD, and t

lag. G. Flux of D-[U-13C]-glucose disposal.

Values represent means ± SEM for n = 5 (ANOVA).

Blood

delay

Lumen

G6Pdirect

adjusted

HK

G6Pase

Disposal

Sampling site

0

1000

2000

3000

0 1 2 3

Fxr +/+

Fxr -/-

Bloo

d D

-[U

-13C]

-glu

cose

(μM

)

Time (hours)

B

0

100

200

0 1 2 3

FxrFxrFxrFxrFxr +/+

Fxr -/-

Dire

ct fl

ux (μ

mol

/kg/

min

)

Time (hours)

C

FxrFxrFxr +/+

Fxr -/-

0

20

40

60

0 1 2 3Time (hours)

Gluc

ose-

6-ph

osph

atas

e flu

x

(μm

ol/k

g/m

in)

E

0

100

200

0 1 2 3

FxrFxrFxr +/+

Fxr -/-

Hex

okin

ase

flux

(μm

ol/k

g/m

in)

Time (hours)

D

0

20

40

60

0 1 2 3

FxrFxrFxrFxrFxr +/+

Fxr -/-

Tota

l flux

(μm

ol/k

g/m

in)

Time (hours)

F

0

20

40

60

0 1 2 3Time (hours)

FxrFxr

D-[U

-13C]

-glu

cose

dis

posa

l

(μm

ol/k

g/m

in)

FxrFxrFxrFxrFxr +/+

Fxr -/-

G

A


DISCUSSIONFXR is a bile acid-activated nuclear receptor that regulates biosynthesis and enterohepatic trans-

port of bile acids [83,84]. Recently, it was shown that FXR mRNA levels and activity are regulated by

glucose [88]. In addition, FXR controls expression of several genes encoding enzymes in gluconeo-

genesis, e.g., Pepck, Fbp1, and G6ph [89]. These fi ndings indicate a role for FXR in control of hepatic

glucose metabolism, particularly during the fasting-feeding transition [44]. Recent reports showed

the presence of FXR in the absorptive epithelium of the small intestine [93]. Results of the current

study clearly show that Fxr -/- mice have delayed intestinal glucose absorption due to an enhanced

G6P turnover in the proximal enterocytes. Thus, these results add an extra regulatory role to FXR in

the regulation of energy substrate metabolism.

We noticed a diff erence in blood glucose increase between wild-type and Fxr -/- mice during an

OGTT (Figure 1). The increase in blood glucose was clearly delayed in Fxr -/- mice and we therefore

speculated that Fxr -/- mice might have 1) an enhanced glucose disposal rate; 2) a reduced and/or

delayed intestinal glucose absorption rate; and/or 3) a less eff ective suppression of endogenous

glucose production upon OGTT. Accordingly, we decided to analyze intestinal glucose absorption

and glucose clearance applying single-pool, fi rst-order kinetics to distinguish between these pos-

sibilities. Isotopic data were used, obtained by OGTT enriched with D-[U-13C]-glucose while the mice

were infused with a trace amount of D-[6,6-2H2]-glucose before and during the OGTT.

Using the blood D-[U-13C]-glucose concentrations solely, glucose absorption and elimination pa-

rameters were estimated. The elimination constant (kel) was signifi cantly lower in Fxr -/- mice. Thus,

compared to wild-type mice, Fxr -/- mice had a signifi cantly increased blood glucose half-life. There-

fore, our fi rst hypothesis to explain the hampered increase in blood glucose upon an oral glucose

bolus in Fxr -/- mice, i.e, an enhanced glucose disposal rate in these mice, has been falsifi ed. Based on

earlier work [45], this outcome was expected.

Using both D-[U-13C]-glucose and D-[6,6-2H2]-glucose data, we were able to calculate glucose

turnovers and intestinal glucose absorption under non-steady state conditions (Figure 3). Compared

to wild-type, Fxr -/- mice had a reduced appearance of glucose in the fi rst 45 minutes, which was

compensated in the period thereafter, resulting in recoveries that were almost the same between

both genotypes at the end of the experiment (Figures 3D). This fi ts with the observation that hardly

any glucose was left in the intestinal lumen at 30 and 180 minutes after oral glucose administration

(Figure 4A). These data establish that Fxr -/- mice have a delayed but not a decreased glucose appear-

ance rate.

Previously, Cariou et al. [44] showed that Fxr -/- mice had a reduced EGP compared to wild-types,

which we could confi rm in our experiments (Figure 3E). The data from the current study suggest

reduced hepatic insulin sensitivity because the suppression of the EGP upon the OGTT was less

pronounced in Fxr -/- mice compared to wild-type mice (Figure 3E). Remarkably, the OGTT-mediated

reductions of the EGP in Fxr -/- and wild-type mice (Figure 3E) were fully comparable with was seen

before when hyperinsulinemic euglycemic clamp experiments were performed [44]. We also found

a tendency towards increased plasma insulin concentrations in the Fxr -/- mice shortly after the OGTT

(Figure 1C). For one hour after the OGTT onwards, the plasma insulin levels were statistically signifi -

cant increased in the Fxr -/- mice. The higher plasma insulin levels of Fxr -/- mice (Figure 1C) coincided

with lower liver glycogen and liver G6P concentrations (Figures 4B and 4C), again pointing towards

5 5Chapter 3

a reduced hepatic insulin sensitivity. In addition, the increased blood glucose MRT (Table 2) in

Fxr -/- mice point towards a reduced peripheral insulin sensitivity, as has also been shown before

[45,90]. Thus, the current and previous studies [44,45,90] show reduced hepatic and peripheral insu-

lin sensitivity in Fxr -/- mice.

From our initial hypotheses to explain the hampered increase in blood glucose during an OGTT

in Fxr -/- mice, only a delayed appearance of intestine-derived glucose in Fxr -/- mice holds. This de-

layed appearance can be explained by a delayed glucose transport through the enterocyte and/

or enhanced absorption of portal glucose by the liver. The latter is unlikely in view of the reduced

hepatic glycogen and G6P concentrations at 30 minutes after the oral glucose dose (Figures 4B and

4C). We developed a compartmental model to simulate the consequences of an enhanced glucose

metabolism inside enterocytes on the kinetics of glucose absorption. The model was based on the

observations published by Stumpel et al. [91]. They showed that in isolated intestines of Glut2-/- mice,

addition of the G6Pase inhibitor S4048 almost completely abolished glucose transport across the

intestinal wall. Apparently, when the direct transport of glucose across the intestinal wall via SGLT1

and GLUT2 is absent, glucose transport proceeds by means of an indirect pathway involving glucose

phosphorylation/dephosphorylation inside enterocytes. When this model is applied using our glu-

cose data, it becomes clear that Fxr -/- mice have an enhanced fl ux through the enterocytic G6P pool

compared to wild-type mice (Figures 6D and 6F). An enhanced enterocyte glucose cycling is sup-

ported by the observation that the oral glucose administration resulted in a 6-fold increase of G6P in

the proximal part of the small intestine in Fxr -/- mice, whereas this increase was absent in wild-type

mice (Figure 4D). The increased Hk1 and Hk2 mRNA levels in the proximal part of the small intestine

in Fxr -/- mice compared to wild-type mice (Figures 5C and 5D) also underscore an increased conver-

sion of glucose in G6P in this part of the intestine.

Remarkably, the largest eff ects of Fxr defi ciency on gene transcription were found in the very

proximal part of the small intestine (Figures 5C and 5D), the part considered not to contribute to ab-

sorption of bile acids secreted into the bile. So, the physiological relevance of bile acids in control of

intestinal glucose metabolism is unclear and needs more investigation. The role of FXR as a glucose-

regulated nuclear transcription factor [88] suggests a physiological function in intestinal glucose

absorption. Whether postprandial bile acids activate FXR in proximal small intestine remains to be

established. The presence of FXR in tissues that normal not exposed to bile acids, e.g., adipose tissue,

adrenal glands, and skin [93], suggests the existence of alternative endogenous FXR ligands.

In conclusion, the experiments described in this paper show that Fxr -/- mice have delayed intes-

tinal glucose absorption, supporting a novel regulatory role of FXR in the enterocyte. Once again,

these studies show that bile acid, carbohydrate, and lipid metabolism are closely linked. In addition,

this paper shows the feasibility of the single pool, fi rst-order kinetic model to study kinetics of intes-

tinal glucose absorption and processing with stable isotopes.

ACKNOWLEDGEMENTSThe authors would like to thank Rick Havinga, Theo Boer, and Gemma Brufau for skillful technical

assistance. This work is supported by EU Grant Hepadip (No.018734), an unrestricted research grant

from Daiichi Sankyo, Inc. (Parsippany, NJ) and grants from the Agence Nationale de la Recherche (No.

A05056GS, No. PPV06217NSA and ANR-06-PHYSIO-027-01, Project R06510NS).

4PPARα activation simultaneously induces hepatic fatty acid oxidation,

synthesis and elongation in mice

M.H. Oosterveer

A. Grefhorst

T.H. van Dijk

H. Havinga

B.Staels

F. Kuipers

A.K. Groen

D-J. Reijngoud

CONDITIONALLY ACCEPTEDFOR PUBLICATION

5 8 PPARα activation induces fatty acid synthesis

ABSTRACTA growing body of evidence indicates that PPARα not merely serves as a transcriptional regulator of

fatty acid catabolism, but exerts a much broader role in hepatic lipid metabolism.

We determined adaptations in hepatic lipid metabolism and related aspects of carbohydrate me-

tabolism upon treatment of C57Bl/6 mice with the PPARα agonist fenofi brate. Stable isotope pro-

cedures were applied to assess hepatic fatty acid synthesis, fatty acid elongation and carbohydrate

metabolism.

Fenofi brate treatment strongly induced hepatic de novo lipogenesis and chain elongation

(+~300%, +~150% and +~600% for C16:0, C18:0 and C18:1 synthesis respectively), in parallel to an

increased expression of lipogenic genes. The lipogenic induction in fenofi brate-treated mice was

found to depend on SREBP-1c but not ChREBP. Fenofi brate treatment resulted in a reduced contri-

bution of glycolysis to acetyl-CoA production, while cycling of G6P through the PPP was presumably

enhanced.

Altogether, our data indicate that β-oxidation and lipogenesis are simultaneously induced upon

PPARα activation. These observations may refl ect a physiological mechanism by which PPARα and

SREBP-1c collectively ensure proper handling of fatty acids to protect the liver against cytotoxic

damage.

5 9Chapter 4

INTRODUCTIONFatty acids are cytotoxic molecules. Both their oxidation and storage as TGs may be important to

protect the liver against lipotoxicity. It was recently postulated that an increased conversion of satu-

rated fatty acids into mono-unsaturated fatty acids (MUFAs), stimulates storage as TG and prevents

NEFA-induced hepatocellular apoptosis [100]. However, the mechanisms underlying this lipogenic

response has remained enigmatic. PPARs represent likely candidates to mediate this response be-

cause these nuclear receptors act as cellular fatty acid sensors.

PPARα induces remodelling of hepatic lipid metabolism under conditions of increased fatty acid

infl ux, such as fasting and high-fat feeding [13,14,101]. Upon activation, PPARα induces the expres-

sion of a multitude of genes encoding proteins involved in peripheral lipid mobilization and fatty

acid oxidation. [13,14,102–104]. In addition PPARα plays a role in hepatic lipid droplet formation

[105,106], and mediates adaptive responses to prevent oxidative stress and the accumulation of

cytotoxic NEFAs [2]. For example, PPARα promotes the degradation of lipid-derived infl ammatory

mediators [107] and induces mitochondrial uncoupling as well as anti-oxidant systems to protect

against oxidative damage associated with (incomplete) β-oxidation [108–114]. As a consequence,

PPARα activity protects against hepatic infl ammation in mice [115–118].

Fibrates are pharmacological PPARα agonists that are clinically used to treat dyslipidemia [49].

Interestingly, PPARα agonist treatment has also been shown to promote 3H2O incorporation into

hepatic lipids in wild-type but not in Pparα-/- mice [119]. This strongly suggests that, besides an in-

crease in hepatic fatty acid oxidation, hepatic fatty acid synthesis is enhanced in response to PPARα

activation. How this observation relates to hepatic MUFA synthesis and TG storage in the liver [100]

remains to be elucidated. In this respect it is interesting to note that SCD1, the lipogenic enzyme

controlling MUFA synthesis, has been reported to be a direct PPARα target gene [120].

Considering the regulatory role of PPARα under conditions of increased fatty acid infl ux, specifi c

changes in hepatic processing of fatty acids are to be expected upon PPARα activation. To gain

insight into these changes, we used sophisticated stable isotope techniques to quantify de novo

lipogenesis and fatty acid elongation in vivo in mice that were treated with the PPARα agonist fenofi -

brate. To evaluate the interactions between hepatic glucose and lipid metabolism, we also deter-

mined relevant hepatic carbohydrate fl uxes.

EXPERIMENTAL PROCEDURESAnimals

To assess the eff ects of PPARα activation on metabolite concentrations and metabolite fl uxes in vivo,

male C57Bl/6 mice (Charles River, L'Arbresle Cedex, France) were housed in a light- and temperature-

controlled facility (lights on 6:30 AM-6:30 PM, 21 °C). They were fed a standard laboratory chow

diet (A03; UAR, Villemoison-sur-Orge, France) with or without fenofi brate (0.2% wt/wt) during two

weeks and had free access to drinking water. Experimental procedures were approved by the Ethics

Committees for Animal Experiments of the University of Groningen. To determine transcriptional

regulation of lipogenic gene expression, female Srebp-1c -/- and Chrebp-/- mice and their wild-type lit-

termates [21,121] were housed in a light- and temperature-controlled facility (lights on 6:00 AM-6:00

PM, 21 °C). They were fed a standard laboratory chow diet (7002, Harlan Teklad Premier Laboratory


Diets, Madison, WI) with or without fenofi brate (0.2% wt/wt) during two weeks and had free access

to drinking water. The experiments involving the Srebp-1c -/- and Chrebp-/- mice were approved by the

Institutional Animal Care and Research Advisory Committee at the University of Texas Southwestern

Medical Center (Dallas, TX).

Metabolite and gene expression analysis

The C57Bl/6 mice were fasted from 6 AM-1 PM with drinking water available and were subsequently

sacrifi ced by cardiac puncture under isofl urane anaesthesia. Srebp-1c -/- and Chrebp-/- mice and their

wild-type littermates were fasted from 7-11 AM with drinking water available and were subsequent-

ly sacrifi ced by isofl urane overdose. Livers were quickly removed, freeze-clamped and stored at -80

°C. Blood was centrifuged (4000xg for 10 minutes at 4 °C) and plasma was stored at -20 °C. Plasma

TG and β-HB concentrations were determined using commercially available kits (Roche Diagnostics,

Mannheim, Germany). Plasma fi broblast growth factor 21 (FGF-21) concentrations were determined

using a mouse radio immunoassay (Phoenix Pharmaceuticals, Burlingame, CA). Frozen liver was ho-

mogenized in ice-cold PBS. Hepatic protein contents were determined according to Lowry et al.

[122]. Hepatic TG and total cholesterol contents were assessed using commercial available kits (Ro-

che Diagnostics, Mannheim, Germany and Wako Chemicals, Neuss, Germany) after lipid extraction

[69]. Hepatic fatty acid composition was analyzed by gas chromatography [123]. Δ9-Desaturation

indices were calculated from the ratios between C16:1 n-7 and C16:0 and C18:1 n-7/n-9 and C18:0,

respectively. Hepatic G6P and glycogen content were determined as described previously [70,71].

RNA was extracted from livers using Tri reagent (Sigma-Aldrich, St. Louis, MO) and cDNA obtained

by reverse transcription was amplifi ed using the appropriate primers and probes. Primer and probe

sequences for 18S, Acc1, ATP binding cassette a1/g1/g5 (Abca1/g1/g5), fatty acid transporter (Cd36),

Cpt1a, Chrebp, Dgat1 and 2, Fas, Gk, G6ph), G6pt, Glut2, Gpat, 3-hydroxy-3-methylglutaryl-Coenzyme

A synthase 2 (Hmgcs2), Lxrα, Pepck, peroxisome proliferator-activated receptor gamma co-activator

1α/β (Pgc-1α/β), Pdk4, Pk, Scd1, Srebp-1c and uncoupling proteins 2 and -3 (Ucp2 and -3) have been

published (www.LabPediatricsRug.nl). The sequences for all other primers and probes are given in

Supplemental Table 1. All mRNA levels were calculated relative to the expression of 18S and normal-

ized for expression levels of control mice.

Determination of de novo lipogenesis and chain elongation in vivo in C57Bl/6 mice

Mice were equipped with a permanent jugular vein catheter [72] and were allowed a recovery pe-

riod of at least three days. On the day of the experiment, the mice were individually housed and

fasted from 6-10 AM. All infusion experiments were performed in conscious, unrestrained mice. A 0.3

M sodium [1-13C]-acetate (99 atom %, Isotec/Sigma-Aldrich) solution was infused via the jugular vein

catheter at an infusion rate of 0.6 mL/h. After 6 hours of infusion, animals were sacrifi ced by cardiac

puncture under isofl urane anaesthesia. Livers were quickly removed, freeze-clamped and stored at

-80 °C. Liver homogenates were prepared in ice-cold PBS and TG fractions were obtained using thin

layer chromatography as previously described [68]. TGs were hydrolyzed in HCl/acetonitrile (1:22 v/v)

for 45 minutes at 100 °C. Fatty acids were extracted in hexane and derivatized for 15 minutes at room

temperature using Br-2,3,4,5,6-pentafl uorobenzyl/acetonitrile/triethanolamine (1:6:2 v/v). Derivati-

zation was stopped by adding HCl and the fatty acid-PFB derivatives were extracted in hexane.

The fatty acid-PFB mass isotopomer distributions were measured using an Agilent 5975 series GC/

MSD (Agilent Technologies, Santa Clara, CA). Gas chromatography was performed using a ZB-1 col-

6 1Chapter 4

umn (Phenomenex, Torrance, CA). Mass spectrometry analysis was performed by electron capture

negative ionization using methane as moderating gas.

The normalized mass isotopomer distributions measured by GC-MS (m0-m

x) were corrected for

natural abundance of 13C by multiple linear regression [39] to obtain the excess fractional distri-

bution of mass isotopomers (M0-M

x) due to incorporation of [1-13C]-acetate. This distribution was

used in MIDA algorithms to calculate the acetyl-CoA precursor pool enrichment (pacetate

), fractional

palmitate synthesis rates (fC16:0

) and the fraction of palmitate and oleate generated by elongation of

de novo synthesized palmitate (fC18:0/1(C16DNL)

), or by elongation of pre-existing palmitate (fC18:0/1(C16PE)

) as

described [124].

In vivo hepatic carbohydrate fl ux measurements in C57Bl/6 mice

Mice were equipped with a permanent jugular vein catheter as described above. After recovery,

the mice were fasted from 6-10 AM. Conscious, unrestrained mice were infused with a solution

containing [U-13C]glucose (7 μM), [2-13C]glycerol (82 μM), [1-2H]galactose (17 μM) and paracetamol

(1 mg/mL) during 6 hours at an infusion rate of 0.6 mL/h as previously described [74]. Blood glu-

cose concentrations were measured every 30 minutes. Blood and urine spots were collected every

60 minutes. Analytical procedures for extraction of glucose from blood spots, derivatization of the

extracted compounds and GC-MS measurements of derivatives were performed according to van

Dijk et al. [40] and corrected for natural abundance of 13C [39]. From this, hepatic carbohydrate fl uxes

were calculated using MIDA as previously described [41]. Supplemental Figure 1 depicts the isotopic

model used.

Statistics

All data are presented as mean values ± SEM. Statistical analysis was performed by Kruskal Wallis/

Mann-Whitney U-test using SPSS for Windows software (SPSS 12.02, Chicago, IL, USA). Statistical

signifi cance was reached at a p value below 0.05.


RESULTSThe catabolic phenotype of fenofi brate-treated mice is accompanied by an induction of hepatic

lipogenic gene expression and accumulation of TGs in the liver

Fenofi brate treatment induced a catabolic phenotype: it caused weight loss without aff ecting food

intake (Table 1). Hepatic Fgf-21 mRNA expression (Table 2) was induced and FGF-21 plasma concen-

trations (Table 1) were increased. Plasma NEFA concentrations did not diff er between fenofi brate-

treated animals and controls (Table 1) while plasma TG concentrations were decreased by fenofi -

brate treatment (Table 1). Fenofi brate treatment induced hepatic peroxisome proliferation which

resulted in increased liver weight and hepatic protein content (Table 1). Hepatic fatty acid oxidation

was promoted by fenofi brate treatment, as indicated by the anticipated increase in expression of

genes involved in fatty acid transport, ketogenesis and peroxisomal and mitochondrial β-oxidation

(Table 2) as well as the elevated plasma β-HB concentrations (Table 1). The expression of genes en-

coding proteins involved in uncoupling of oxidative phosphorylation was also increased (Table 2).

The catabolic phenotype of fenofi brate-treated mice was associated with an increase in hepat-

ic TG content while cholesterol content remained unaff ected (Table 1). The hepatic expression of

genes encoding enzymes involved in de novo lipogenesis, e.g., Acc1 and Fas was higher in these

animals while that of Srebp-1c and its co-activator Pgc-1β remained unaltered (Table 2). In addi-

tion, expression of genes encoding enzymes involved in fatty acid elongation and desaturation e.g.,

Elovl5, Scd1, Fads1 and Fads2, was markedly induced upon fenofi brate treatment (Table 2). Changes

in hepatic fatty acid synthesis and elongation/desaturation gene expression pattern translated into

altered hepatic fatty acid composition (Table 3) with a marked increase in the abundance of MUFA,

resulting in increased hepatic Δ9-desaturation indices (Table 1).

Table 1. General characteristics, plasma and hepatic metabolite levels.

control fenofi brate

Body weight change (%) 7.8±1.0 -6.4±2.0*

Food intake (g/day) 4.5±0.1 4.8±0.4

Plasma FGF-21 (ng/mL) 1.1±0.1 4.7±0.6*

Plasma NEFA (mmol/L) 0.27±0.04 0.20±0.01

Plasma β-HB (mmol/L) 0.15±0.03 1.00±0.12*

Plasma TGs (mmol/L) 0.53±0.04 0.10±0.01*

Liver weight (% of body weight) 4.8±0.1 13.2±0.5*

Hepatic protein (mg/g) 160±3 171±2*

Hepatic C16 desaturation index 0.07±0.01 0.13±0.02*

Hepatic C18 desaturation index 1.20±0.10 4.06±0.20*

Hepatic TGs (μmol/g) 15.4±2.6 24.6±2.0*

Hepatic cholesterol (μmol/g) 7.7±0.4 8.2±0.3

Values represent means ± SEM for n=6; * p<0.05 fenofi brate vs. control (Mann-Whitney U-test).

6 3Chapter 4

Massive induction of the lipogenic fl ux contributes to hepatic TG accumulation in fenofi brate-treated

mice

To establish the physiological relevance of the induction in lipogenic gene expression, we deter-

mined de novo lipogenesis and fatty acid elongation and their contributions to hepatic TG in vivo.

We therefore infused 1-13C acetate into mice for 6 hours and applied MIDA to diff erentiate between

de novo lipogenesis and fatty acid elongation as described [124]. Fenofi brate treatment resulted in

a massive increase in de novo lipogenesis (Figure 1A). Elongation of both de novo synthesized and

pre-existing palmitate was also higher in fenofi brate-treated mice, which resulted in an increase in

stearate and oleate synthesis (Figure 1B and 1C). These results are consistent with the increased ex-

pression of genes encoding enzymes involved in fatty acid synthesis, elongation, and desaturation

upon fenofi brate treatment (Table 1). Moreover, the results from the isotope infusion studies explain

the changes observed in the hepatic fatty acid profi le (Table 3), i.e., the higher desaturation indices

and the ~50% increase in oleate content. Interestingly, the synthesis rates of TG-associated palmi-

tate, oleate and stearate (Figure 1A-C) were very similar to the values observed for the total hepatic

synthetic rates of these fatty acids. Hence, the acetyl-CoA precursor pool enrichments in total and

TG-associated palmitate were similar. Strikingly, the acetyl-CoA precursor pool enrichment remained

unaff ected upon fenofi brate treatment (Figure 1D), indicative for a similar and rapid turnover of the

acetyl-CoA precursor pool.

total TG0

10

20

30

40controlfenofibrate

Frac

tiona

l C16

:0 s

ynth

esis

(%)

* *

total TG0

10

20

30

40control C16:0 PEfenofibrate C16:0 PE

Frac

tiona

l C18

:1 s

ynth

esis

(%)

* ***

**

control C16:0 DNLfenofibrate C16:0 DNL

total TG0

10

20

30

40control C16:0 PEfenofibrate C16:0 PE

Frac

tiona

l C18

:0 s

ynth

esis

(%) control C16:0 DNL

fenofibrate C16:0 DNL

**

*

*

*

*

total C16:0 TG C16:00

5

10

15

20 controlfenofibrate

Acet

yl-C

oA p

ool e

nric

hmen

t (%

)

A B

C D

Figure 1. Hepatic fatty acid synthesis in control and fenofi brate-treated mice.

A, Fractional synthesis rates of total and TG-derived palmitate from de novo lipogenesis. B, Fractional synthesis rates of total and TG-de-

rived stearate from elongation of labeled (de novo synthesized, C16:0DNL) and unlabeled (pre-existing; C16:0PE) palmitate. C, Fractional

synthesis rates of total and TG-derived oleate from elongation of de labeled (de novo synthesized, C16:0DNL) and (C16:0PE) unlabeled

palmitate. D, Acetyl-CoA precursor pool enrichments in total and TG-derived palmitate.

Open bars, control group; fi lled bars, fenofi brate-treated group. Values represent means ± SEM for n=6-8; * p<0.05 fenofi brate vs. control



Table 2. Hepatic mRNA expression levels.


Fatty acid mobilization/uptake

Fgf-21 1.0±0.3 4.4±0.4*

Cd36 1.0±0.1 7.2±0.7*

β-oxidation and keto+genesis

Aox 1.0±0.1 4.5±0.5*

Cpt-1a 1.0±0.1 1.4±0.1*

Lcad 1.0±0.1 2.6±0.2*

Hmgcs2 1.0±0.1 1.9±0.2*

Mitochondrial uncoupling

Ucp2 1.0±0.1 2.9±0.2*

Ucp3 1.0±0.2 109.8±9.6*

Fatty acid synthesis

Srebp-1c 1.0±0.1 1.3±0.1

Pgc-1ß 1.0±0.1 1.0±0.1

Acc1 1.0±0.1 1.7±0.1*

Fas 1.0±0.1 1.5±0.1*

Elovl6 1.0±0.2 1.3±0.1

Scd1 1.0±0.1 2.6±0.2*

Elovl5 1.0±0.1 3.2±0.3*

Fads1 1.0±0.1 1.5±0.1*

Fads2 1.0±0.2 2.7±0.3*

Glucose uptake/glycolysis

Chrebp 1.0±0.1 0.7±0.1*

Glut2 1.0±0.1 0.5±0.1*

Gk 1.0±0.1 0.4±0.0*

Pk 1.0±0.1 0.2±0.0*

Pdk4 1.0±0.3 27.0±2.8*

Gluconeogenesis

Pgc-1α 1.0±0.1 1.1±0.1

Pepck 1.0±0.1 0.8±0.1

G6ph 1.0±0.3 0.8±0.1

G6pt 1.0±0.0 0.5±0.2*

Gyk 1.0±0.1 1.8±0.1*

PPP and NADPH synthesis

G6pdh 1.0±0.1 1.0±0.1

6Pdgh 1.0±0.0 2.2±0.1*

Taldo1 1.0±0.1 1.8±0.2*

Tkt 1.0±0.1 1.2±0.1

Me1 1.0±0.1 5.9±0.5*

Expression levels were normalized to 18S expression and values represent means ± SEM for n=6; * p<0.05 fenofi brate vs. control (Mann-

Whitney U-test).

6 5Chapter 4

Table 3. Hepatic fatty acid profi les.


C14:0 0.24±0.03 0.27±0.01

C16:1 n-7 1.76±0.22 4.55±0.56*

C16:0 24.04±0.98 34.77±0.95*

C18:3 n-6 0.23±0.02 0.26±0.01

C18:2 n-6 20.69±1.05 16.09±0.67*

C18:3 n-3 0.62±0.07 0.26±0.04*

C18:1 n-9 12.76±1.28 28.57±1.21*

C18:1 n-7 2.07±0.13 5.79±0.20*

C18:0 12.32±0.34 8.50±0.30*

C20:4 n-6 11.44±0.26 10.82±0.36

C20:5 n-3 0.70±0.03 0.51±0.03*

C20:3 n-9 0.14±0.01 0.61±0.02*

C20:3 n-6 1.12±0.04 5.43±0.22*

C20:2 n-6 0.33±0.02 0.42±0.02*

C20:1 n-9 0.36±0.03 0.55±0.02*

C20:0 0.20±0.01 0.08±0.00*

C22:5 n-6 0.13±0.01 0.14±0.01

C22:6 n-3 7.91±0.26 6.72±0.16*

C22:4 n-6 0.21±0.01 0.41±0.02*

C22:5 n-3 0.58±0.03 0.85±0.04*

C22:0 0.44±0.02 0.18±0.01*

C24:1 n-9 0.48±0.01 0.39±0.01*

C24:0 0.37±0.01 0.14±0.00*

Values represent means ± SEM for n=6 and expressed in μmol/g liver; * p<0.05 fenofi brate vs. control (Mann-Whitney U-test).

The induction of lipogenic genes upon fenofi brate treatment is transcriptionally regulated by

SREBP-1c

PPARα agonist treatment has been reported to be associated with an increased abundance of nu-

clear SREBP-1c [119], which might be responsible for the observed induction in lipogenic gene ex-

pression. On the other hand, lipogenic gene expression might be transcriptionally controlled by

ChREBP [125]. To assess whether the induction of the lipogenic genes upon fenofi brate treatment

depended on SREBP-1c and/or ChREBP, Srebp-1c -/- and Chrebp-/- mice were treated with fenofi brate.

Figure 2 shows the expression profi les of genes encoding enzymes controlling de novo lipogenesis

as well as fatty acid elongation and desaturation in the knockout mice and their wild-type litterma-

tes. Compared to their wild-type littermates, the induction of fatty acid synthesis genes upon fenofi -

brate treatment was clearly blunted in Srebp-1c -/- mice. In Chrebp-/- mice, however, the induction

of these genes was maintained. Similarly, induction of genes encoding enzymes controlling fatty

acid esterifi cation as well as nicotinamide adenine dinucleotide phosphate (NADPH) synthesis was

blunted in Srebp-1c -/- mice only (Table 4).



Srebp-1c +/+ control Srebp-1c +/+ fenofi brate Srebp-1c -/- control Srebp-1c -/- fenofi brate

Fatty acid esterifi cation

Dgat1 1.0±0.1 2.4±0.1* 1.0±0.0 1.3±0.1

Dgat2 1.0±0.0 0.8±0.0 1.0±0.0 0.6±0.0*

Gpat 1.0±0.0 1.8±0.0* 1.0±0.0 1.2±0.1

PPP/NADPH synthesis

G6pdh 1.0±0.1 2.0±0.5 1.0±0.2 0.7±0.1

6Pdgh 1.0±0.1 3.2±0.2* 1.0±0.0 1.5±0.2*

Taldo1 1.0±0.1 1.5±0.1* 1.0±0.1 1.1±0.0

Me1 1.0±0.1 6.0±0.4* 1.0±0.1 2.8±0.3*

Chrebp +/+ control Chrebp +/+ fenofi brate Chrebp -/- control Chrebp -/- fenofi brate

Fatty acid esterifi cation

Dgat1 1.0±0.1 2.6±0.3* 1.0±0.0 3.0±0.1*

Dgat2 1.0±0.1 0.6±0.1* 1.0±0.0 0.8±0.0

Gpat 1.0±0.1 1.3±0.1 1.0±0.0 2.0±0.1*

PPP/NADPH synthesis

G6pdh 1.0±0.1 0.9±0.2 1.0±0.1 1.2±0.1

6Pdgh 1.0±0.0 2.4±0.3* 1.0±0.1 3.3±0.3*

Taldo1 1.0±0.1 1.8±0.3* 1.0±0.1 1.8±0.3*

Me1 1.0±0.1 4.9±0.4* 1.0±0.1 10.8±1.2*

Expression levels were normalized to 18S expression. Values of untreated mice of each genotype were set to 1.

Values represent means ± SEM for n=4; * p<0.05 fenofi brate vs. control and # p<0.05 knockout vs. wild-type (Mann-Whitney U-test).

Acc1 FasElovl6 Scd1

Elovl5 Fads1

Fads2

0

2

4

6

8

10

12 Srebp-1c +/+ controlSrebp-1c +/+ fenofibrateSrebp-1c -/- controlSrebp-1c -/- fenofibrate

Rela

tive

mRN

A ex

pres

sion

**

*

*

* *

Acc1 FasElovl6 Scd1

Elovl5 Fads1

Fads2

0

2

4

6

8

10

12 Chrebp +/+ controlChrebp +/+ fenofibrateChrebp -/- controlChrebp -/- fenofibrate

Rela

tive

mRN

A ex

pres

sion

** *

*

*

*

* * **

* ** *

A B

Figure 2. Transcriptional control of lipogenic gene expression in control and fenofi brate-treated mice.

Gene expression levels were normalized for 18S expression. Expression levels of untreated mice of each genotype were set to 1. A,

Expression of genes involved in fatty acid synthesis in Srebp-1c -/- mice and wild-type littermates. B, Expression of genes involved in fatty

acid synthesis in Chrebp-/- mice and wild-type littermates.

Open bars, wild-type control group; fi lled bars, wild-type fenofi brate-treated group; dashed bars, knockout control group; dotted bars,

knockout fenofi brate-treated group. Values represent means ± SEM for n=4; * p<0.05 fenofi brate vs. control (Mann-Whitney U-test).

6 7Chapter 4

Besides SREBP-1c and ChREBP, LXR is another important transcriptional regulator of lipogenic genes.

Expression analysis of Lxrα and its direct target genes Abca1, Abcg5 and Abcg1 did not provide evi-

dence for an increased LXR activity upon fenofi brate treatment (Table 5).



Lxrα 1.0±0.0 0.9±0.1

Abca1 1.0±0.2 1.1±0.1

Abcg5 1.0±0.2 0.8±0.2

Abcg1 1.0±0.2 0.9±0.1

Values represent means ± SEM for n=6.

Reduced hepatic glucose consumption upon fenofi brate treatment is compensated for by an

increased gluconeogenic fl ux and enhanced PPP cycling

Glucose provides acetyl-CoA required for fatty acid synthesis via the glycolytic pathway. To assess

the contribution of adaptations in hepatic glucose metabolism to the increased lipogenic fl ux, we

determined carbohydrate fl uxes in vivo by MIDA following isotope infusion as described [40,41]. The

isotopic model used is depicted in Supplemental Figure 1 and the primary isotopic parameters are

listed in Table 6. Blood glucose concentrations were comparable during isotope infusion in both

groups (Table 6). Fenofi brate treatment resulted in a lower hepatic glucose uptake, indicated by a

decreased fl ux through glucokinase (Figure 3A). This was paralleled by a decreased hepatic mRNA

expression of Chrebp, Glut2 and Gk (Table 2). In addition, the decreased expression of hepatic Pk

mRNA, encoding a key enzyme in glycolysis, and the massive induction of Pdk4 mRNA, encoding the

major inhibitor of glycolysis (Table 2) indicate a reduced glycolysis upon fenofi brate treatment.

Table 6. Blood glucose concentrations and primary isotopic parameters during steady-state infusion (180-360 min).


Blood glucose (mM) 8.2±0.2 8.5±0.6

Isotope dilution

d(glc) 0.018±0.001 0.018±0.002

d(UDPglc) 0.196±0.010 0.166±0.009*

Isotope exchange

c(glc) 0.176±0.011 0.079±0.005*

c(UDPglc) 0.137±0.008 0.157±0.008

MIDA analysis

f(glc) 0.55±0.02 0.59±0.02

f(UDPglc) 0.46±0.02 0.53±0.01*

Values represent means ± SEM for n=5-6; * p<0.05 fenofi brate vs. control (ANOVA).


0

200

400

600


G6P

(nm

ol/g

)

*

0

100

200

300


Glyc

ogen

(μm

ol/g

)

*

A B

Figure 4. Hepatic G6P and glycogen content in control and fenofi brate-treated mice.

Open bars, control group; fi lled bars, fenofi brate-treated group. Values represent means ± SEM for n=6; * p<0.05 fenofi brate vs. control


Figure 3. Hepatic glucose metabolism in control and fenofi brate-treated mice during steady-state infusion (t= 180-360 min).

A, Glucokinase fl ux. B, Gluconeogenic fl ux and partitioning towards glucose (light grey bars) and UDP-glucose (dark grey bars). C, Hepa-

tic glucose production rate and contribution of gluconeogenic fl ux (light grey bars) and glucose cycling (dark grey bars). D, Glycogen

phosphorylase fl ux. E, Glycogen balance and F, Abundance of triply labeled molecules in blood and UDP-glucose.

Open bars, control group; fi lled bars, fenofi brate-treated group. Values represent means ± SEM for n=5-6; * p<0.05 fenofi brate vs. control

(ANOVA).

0

10

20

30

40


Gluc

okin

ase

flux

(μm

ol/k

g/m

in)

*

A

0

50

100


Glyc

ogen

pho

spho

rlase

flux

(μm

ol/k

g/m

in)

D

-80

-60

-40

-20

0


Glyc

ogen

bal

ance

(μm

ol/k

g/m

in) E

control fenofibrate0

50

100

150GNG (glucose)GNG (UDP-glucose)

Gluc

oneo

geni

c flu

x(μ

mol

/kg/

min

)

**

B

control fenofibrate0

100

200

300Ra (glc;endo)r (glc)

Gluc

ose

prod

uctio

n(μ

mol

/kg/

min

)

*

C

blood glucose UDP-glucose0.0

0.2

0.4

0.6

0.8

1.0 controlfenofibrate

M3 a

bund

ance

(%)

* *

F

6 9Chapter 4

The reduction in hepatic glucose input from the circulation was compensated for by an increased

de novo synthesis of G6P, i.e., an increased gluconeogenic fl ux (Figure 3B). Expression of Gyk, which

encodes the enzyme that facilitates the use of glycerol as a gluconeogenic substrate, was also in-

duced upon fenofi brate treatment. On the other hand, expression of other gluconeogenic genes,

e.g., Pgc-1α, Pepck and G6ph remained unaff ected while that of G6pt was reduced upon fenofi brate

treatment (Table 2). The increased gluconeogenic fl ux in fenofi brate-treated mice did not promote

hepatic glucose production: hepatic glucose output (i.e., the G6Pase fl ux) was hardly aff ected by

fenofi brate treatment. This is explained by the lower compartmentation of the gluconeogenic fl ux

towards blood glucose (86±1% vs. 80±1%, control vs. fenofi brate, p<0.05) and the reduced glucose

cycling (from blood glucose to G6P back to blood glucose, Figure 3C) in fenofi brate-treated mice.

Moreover, the increased gluconeogenic fl ux did not increase glycogen disposition because of an

increased fl ux through glycogen phosphorylase (Figure 3D and 3E). Hepatic G6P and glycogen con-

tent were even reduced in fenofi brate-treated mice (Figure 4). The increased expression of genes

encoding enzymes that mediate the PPP 6-phosphogluconate dehydrogenase and transaldolase 1

(6Pdgh and Taldo1, Table 2) as well as the higher abundance of triple-labeled glucose molecules in

the UDP- and blood glucose pools (Figure 3F) strongly suggest an enhanced fl ux through the PPP

upon fenofi brate treatment. Cycling through the PPP generates NADPH, which is needed to main-

tain the increased energy-consuming lipogenic fl ux in fenofi brate-treated animals. This is consistent

with the induction of hepatic malic enzyme 1 (Me1) expression upon fenofi brate treatment (Table

2). Me1 encodes another NADPH-generating enzyme, and its expression has been reported to be

controlled by PPARα [126].


DISCUSSIONThe current study is the fi rst to establish the remodelling of hepatic intermediary metabolism that

occurs upon chronic PPARα activation in mice. In parallel to the well-known increase in hepatic

fatty acid oxidation, fenofi brate treatment resulted in a massive induction of the lipogenic fl ux and

a concomitant adjustment of hepatic glucose metabolism to provide NADPH required for fatty acid

synthesis and the subsequent esterifi cation to form TG. In addition, we show that fenofi brate treat-

ment reduced glycolysis and thus acetyl-CoA supply from glucose. Altogether, these data provide

evidence for the existence of an adaptive response to an increased fatty acid infl ux and catabolism

that will protect the liver against the toxic eff ects of excess intracellular NEFA and their oxidation

products.

PPARα action was originally found to be crucial for the hepatic adaptive response to fasting [13,14].

Increased PPARα activity enhances the fl ux of fatty acids from the adipose tissue to the liver via the

action of FGF-21. Hepatic Fgf-21 is a direct target gene of PPARα, and both fasting and pharmacologi-

cal PPARα activation result in an increase in circulating FGF-21 concentrations [103,104]. FGF-21 in

turn acts directly on adipose tissue to stimulate lipolysis [103]. We observed a 4.5-fold induction of

both hepatic Fgf-21 expression and its plasma concentration (Table 1/2) upon fenofi brate treatment.

The hepatic expression of Cd36, a major fatty acid transporter, was concomitantly induced (Table 2).

Thus, the increased hepatic infl ux of adipose tissue-derived fatty acids was compensated for by an

increased hepatic uptake. As a consequence, circulating NEFA concentrations were maintained.

PPARα agonist treatment has however also been shown to promote 3H2O incorporation into

hepatic lipids in wild-type but not in Pparα-/- mice [119]. These observations suggested a PPARα-

dependent induction of hepatic fatty acid synthesis. However, H2O is used in multiple metabolic

pathways and the results with 3H2O therefore do not truly refl ect the lipogenic fl ux, particularly since

fatty acid oxidation may also contribute to 3H abundance in fatty acids. Moreover, the contributions

of de novo lipogenesis and fatty acid elongation were not established. Finally, the 3H2O study did not

address the relationship between MUFA synthesis and hepatic TG storage, which is of particular in-

terest since these processes have been reported to protect against lipotoxicity [100]. Using 13C-ace-

tate and MIDA, we now show that acetyl-CoA incorporation into the hepatic fatty acids was strongly

induced in fenofi brate-treated mice, indicated by an induction of both de novo lipogenesis and fatty

acid elongation (Figure 1). Although quantitative data on the rate of fatty acid catabolism are cur-

rently not available, the increased hepatic content of the major TG-derived fatty acids indicates that

the rate of fatty acid oxidation was not suffi cient to counterbalance the NEFA infl ux and synthesis.

To delineate the transcriptional mechanism by which fenofi brate treatment increased the lipo-

genic program, we performed gene expression analysis in Srebp-1c -/- and Chrebp-/- mice. As shown in

Figure 2, the lipogenic gene induction was completely accounted for by SREBP-1c, which supports

previous work [119]. ChREBP appears not to be involved since the fenofi brate-mediated induction of

lipogenic genes was not aff ected in Chrebp-/- mice. Expression of Pk, a direct target gene of ChREBP,

was actually found to be strongly reduced upon fenofi brate treatment (Table 2). The question arises

how an increased PPARα activity enhances SREBP-1c mediated gene transcription. Under normal

physiological conditions PPARα and SREBP-1c act in opposite manner. However, the presence of

Pparα has been shown to be required for proper SREBP-1c functioning [127], while our current ob-

servations indicate that Srebp-1c is needed for the induction of Scd1 upon fenofi brate treatment.

7 1Chapter 4

This is surprising since Scd1 has been identifi ed as a direct PPARα target [120]. The relationship

between PPARα and SREBP-1c action therefore requires further investigation, particularly because de

novo lipogenesis generates endogenous PPARα ligands [128].

Because Srebp-1c is a target of LXR [129], and fenofi brate has recently been shown to enhance

LXR promoter activity in vivo we determined the expression of direct LXR target genes in the livers

of fenofi brate-treated mice. We did not observe any induction of Abcg5, Abca1 or Abcg1, which ar-

gues against the involvement of LXR (Table 5). The increased SREBP-1c activity may therefore rather

be related to changes in the intracellular lipid status secondary to an enhanced fatty acid infl ux.

SCD1 action may be crucial in this process [130]. PPARα agonist treatment actually inhibits SREBP-

1c activity and TG synthesis in vitro [131], when the fatty acid fl ux is absent. This strongly suggests

that a fenofi brate-mediated increase in lipolysis and fatty acid infl ux are required for the induction

of lipogenesis in vivo. Indeed, pharmacological PPARα agonists fail to induce both Fgf-21 [132] and

lipogenic genes [133] in the livers of Pparα-/- mice.

We found that both de novo lipogenesis and fatty acid elongation contributed to the increased

lipogenesis upon fenofi brate treatment. These processes require acetyl-CoA, pre-existing palmitic

acid and NADPH. In mice, fenofi brate treatment induces both peroxisomal and mitochondrial fatty

acid oxidation, processes that generate acetyl-CoA and NADPH. Although humans appear to be

resistant to the induction of peroxisome proliferation by PPARα agonists, an increased expression

of hepatic lipogenic genes is also observed in mice that express human PPARα [133]. This lipogenic

induction is therefore most likely related to an elevated mitochondrial fatty acid oxidation driven by

the increased hepatic infl ux and uptake of fatty acids upon fenofi brate treatment. Cytosolic acetyl-

CoA from peroxisomal β-oxidation has been shown to promote fatty acid synthesis via chain elon-

gation [134] while mitochondrial acetyl-CoA is used for ketogenesis and citrate synthesis. Citrate

consumption by the TCA cycle is inhibited by increased NADH levels. As a consequence, citrate is

shuttled from the mitochondria into the cytosol, where it is converted into oxaloacetate and acetyl-

CoA. This acetyl-CoA is used for de novo lipogenesis and chain elongation [134,135], while oxaloace-

tate facilitates transport of acetyl-CoA across the mitochondrial membrane via the pyruvate/malate

cycle, thereby generating NADPH that is required for fatty acid synthesis. Enhanced cycling through

this pathway is evident from the ~6-fold induction of Me1 expression (Table 2). ME1 converts malate

generated from oxaloacetate into pyruvate, hence closing the pyruvate/malate cycle. The expres-

sion of 6Pdgh, encoding another NADPH-generating enzyme, was also increased upon fenofi brate

treatment (Table 2).

Although fenofi brate treatment resulted in a massive induction of the lipogenic fl ux, we observed

comparable acetyl-CoA precursor pool enrichments in treated and untreated mice (Figure 1D), indi-

cating similar acetyl-CoA turnover in both groups. However, the acetyl-CoA input from peroxisomal

and mitochondrial β-oxidation must have been increased in fenofi brate-treated mice. Therefore,

the acetyl-CoA input from other pathways must have been reduced. Hepatic glucose metabolism

provides a major source of hepatic acetyl-CoA and we therefore determined hepatic carbohydrate

fl uxes. In fenofi brate-treated mice, the glucokinase fl ux was reduced by 45% while Pdk4 expression

was ~30-fold induced, indicating reduced hepatic glucose uptake and glycolysis (Figure 3A/Table

2). This presumably refl ects ‘glucose sparing’ [136], and strongly suggests reduced acetyl-CoA supply

from glycolysis. Altogether, this explains the maintenance of acetyl-CoA precursor pool enrichment


in the face of increased β-oxidation upon fenofi brate treatment.

The fl ux through the gluconeogenic pathway was increased upon fenofi brate treatment, as in-

dicated by the enhanced glycerol incorporation (Figure 3B), which supports previous work [137].

The gluconeogenic induction did not result in an increased hepatic glucose output (Figure 3C).

Moreover, the increased gluconeogenic fl ux towards glycogen was balanced by increased glyco-

gen breakdown (Figure 3D/E). Therefore, G6P must have been metabolized via pathways other than

those covered by the isotopic model applied, particularly because hepatic G6P content was reduced

by ~50%. Increased cycling through the PPP upon fenofi brate treatment seems the obvious expla-

nation because expression of 6Pdgh and Taldo1 was induced in the livers of fenofi brate-treated mice

(Table 2). Furthermore, the abundance of triple labeled molecules in the isotopomer patterns of

both blood glucose and UDP-glucose was increased (Figure 3F). We infused U-13C glucose and the

M3 abundance can be considered as a measure of the futile cycling of substrates from G6P through

the PPP back to G6P. Therefore, the higher M3 abundance observed in the blood glucose and UDP-

glucose pools of fenofi brate-treated mice refl ects an increased contribution of triose-phosphate

conversion and hence suggests an enhanced cycling through the PPP. Interestingly, the fl ux through

the PPP has been shown to be reduced in Pparα-/- mice [138]. PPP activity has also been reported

to be increased when the fl ux through glycerol kinase is enhanced [139], as is the case in our expe-

riments. The PPP remodelling of hepatic glucose metabolism upon PPARα activation may not only

provide NADPH needed to maintain the lipogenic fl ux, but may also support anti-oxidant action

since the PPP is coupled to the synthesis of reduced glutathione [140].

In conclusion, we show that fatty acid synthesis and esterifi cation are promoted in response to an

increased fatty acid infl ux and catabolism upon PPARα activation. The presence of SREBP-1c appears

to be essential in this adaptive process. These results add to the growing body of evidence that

PPARα not merely acts as a transcriptional activator of fatty catabolism, but exerts a much broader

role in hepatic lipid metabolism. PPARα agonism also induces remodelling of hepatic glucose me-

tabolism, which presumably serves to support the increased lipogenic fl ux. These metabolic chang-

es are depicted and highlighted in Figure 5. The fl ux through the PPP is increased to supply NADPH

needed for lipogenesis and fatty acid elongation while glycolysis is reduced to prevent excessive

input of acetyl-CoA.

Altogether, our data reveal a novel physiological mechanism by which the liver ensures proper

handling of surplus fatty acids and β-oxidation products to protect itself against lipotoxicity.

ACKNOWLEDGEMENTSThe experiments involving the Srebp-1c -/- and Chrebp-/- mice and their wild-type littermates were

performed in the laboratory of Dr. Jay D. Horton at the Department of Molecular Genetics of the

University of Texas Southwestern Medical Center at Dallas, TX.

The authors thank Dr. Jay D. Horton and Dr. Kosuka Uyeda for providing the Srebp-1c -/- and Chrebp-/-

mice, respectively. The authors thank Theo Boer, Trijnie Bos, Tineke Jager and Frank Perton for excel-

lent technical assistance.

7 3Chapter 4

acetyl-CoA

malonyl-CoA

acetyl-CoA

pyruvate

pyruvate

glucose g-6-p

triose phosphate

glycogen

ketone bodies

citrate

malate

oxaloacetatefatty acids

glycerol

fatty acid-CoAfatty acids

NADPH

NADPH

palmitic acid

triglycerides

acetyl-CoAPEROXISOME

MITOCHONDRION

Figure 5. Remodelling of hepatic intermediary metabolism in fenofi brate-treated mice.

Fenofi brate treatment promotes adipose tissue lipolysis, thereby enhancing hepatic infl ux of glycerol and fatty acids. In the liver, feno-

fi brate promotes fatty acid β-oxidation in peroxisomes and mitochondria. Acetyl-CoA generated by β-oxidation is used for ketogenesis

and energy supply, but also serves as a substrate for fatty acid synthesis via de novo lipogenesis and fatty acid elongation. Acetyl-CoA

transport over the mitochondrial membrane is facilitated by increased pyruvate/malate cycling, which generates NADPH to support the

lipogenic fl ux. In parallel, hepatic glucose uptake and glycolysis are suppressed, and the contribution of acetyl-CoA from hepatic glucose

metabolism to the lipogenic fl ux is consequently reduced. Glycerol is converted into G6P via the gluconeogenic pathway. G6P cycles

through the PPP to triose phosphate, and back to G6P, thereby generating NADPH.

5Fish oil potentiates high fatdiet-induced peripheral

insulin resistance in mice

M.H. Oosterveer

M. Schreurs

T.H. van Dijk

H. Wolters

R. Havinga

S.A.A. van den Berg

K. Willems-van Dijk

G.C.M. van der Zon

D.M. Ouwens

A.K. Groen

F. Kuipers

D-J. Reijngoud

SUBMITTED

7 6 Increased fat oxidation does not improve glucose tolerance

ABSTRACTConfl icting data have been reported concerning the eff ects of fi sh oil on insulin resistance and type

2 diabetes. We have evaluated the metabolic consequences of fi sh oil with regard to substrate uti-

lization and related these to quantitative changes in glucose metabolism in diet-induced insulin

resistant mice.

C57Bl/6 mice were fed high-fat diets containing beef fat or beef fat/fi sh oil and compared to ani-

mals receiving low-fat chow. Whole-body substrate utilization and energy expenditure were deter-

mined by indirect calorimetry. Glucose metabolism was evaluated by stable isotope procedures.

Fish oil decreased the respiratory exchange ratio (RER) in mice fed a high-fat diet while energy

expenditure remained unaff ected. Furthermore, fi sh oil impaired peripheral glucose clearance. Fish

oil decreased basal hepatic glucose production and normalized insulin-clamped hepatic glucose

production. Both high-fat and high-fat/fi sh oil attenuated the insulin-mediated increase in PI3K ac-

tivity in liver and fat.

In conclusion, fi sh oil increases the fat-to-carbohydrate oxidation ratio but does not enhance en-

ergy expenditure. This is associated with a deterioration of high-fat diet-induced adiposity and pe-

ripheral insulin resistance. These data demonstrate that increased fat oxidation alone is not suffi cient

to prevent high-fat diet-induced peripheral insulin resistance in mice.

7 7Chapter 5

INTRODUCTIONIntake of hypercaloric high-fat diets is associated with dyslipidemia, increased cardiovascular risk and

insulin resistance in humans [141]. Although fi sh consumption exerts benefi cial eff ects on blood

lipid profi les and cardiovascular risk [142,143], its consequences for the development of insulin re-

sistance are not conclusive: both improvement and deterioration have been reported [144–150].

Deeper insight into the metabolic adaptations that occur in response to fi sh consumption may help

to explain these discrepant fi ndings. Such information is required to establish the potential of fi sh oil

supplementation for the prevention of insulin resistance.

n-3 PUFA, the bioactive components of fi sh oil, alter the activity of several transcriptional regula-

tors such as PPARs, LXRs, LXRs, and SREBP-1c and ChREBP [33,151,152]. In this way, fi sh oil suppresses

the expression of genes encoding enzymes involved in fat synthesis, while it increases the expres-

sion of fat oxidation enzymes. We have recently shown that fi sh oil replacement of a high-fat diet

inhibits hepatic lipogenesis and VLDL secretion in vivo [124]. The eff ect of fi sh oil replacement on in

vivo substrate oxidation is, however, largely unknown. This is of particular interest considering recent

animal studies showing that high fat oxidation rates are associated with insulin resistance [153–157]

and that glucose disposal is increased when fat oxidation is suppressed [74,153,158,159].

We have therefore evaluated the consequences of dietary fi sh oil for substrate utilization in mice

fed a high-fat diet and related these to alterations in glucose metabolism. C57Bl/6 mice were sub-

jected to a 6-week dietary challenge of a diet rich in beef high-fat or a similar diet in which part of

the fat was replaced by fi sh oil. These high fat-fed mice were compared to animals that received

standard low-fat laboratory chow. We assessed whole-body substrate utilization, energy expendi-

ture and basal and hyperinsulinemic glucose metabolism in vivo by dedicated techniques and re-

lated outcome to biometric and biochemical parameters as well as gene expression patterns and

downstream insulin receptor signalling.

EXPERIMENTAL PROCEDURESAnimals and experimental design

Male C57Bl/6 mice (Charles River, L’Arbresle Cedex, France), three months of age, were housed in a

light- and temperature-controlled facility (lights on 6:30 AM-6:30 PM, 21 °C). They were divided into

groups and fed three diff erent diets for six weeks. One group received laboratory chow (RMH-B, 13

energy% fat), the second group received high-fat diet (beef high-fat, 60 energy% fat) and the third

group received a diet in which part of the high-fat was replaced by fi sh oil (high-fat: 35, fi sh oil: 25 en-

ergy% fat). The two high-fat diets were hypercaloric compared to laboratory chow (chow, 3.3; high-

fat, 5.5; high-fat/fi sh oil, 5.5 kcal/g). All diets were obtained from Abdiets, Woerden, The Netherlands.

For dietary fatty acid composition see Supplemental Table 4. At the end of the dietary period, mice

were either sacrifi ced for basal plasma and tissue collection, subjected to indirect calorimetry, to in

vivo measurements of glucose metabolism, or to evaluation of the IR signalling pathway. Experimen-

tal procedures were approved by the Ethics Committees for Animal Experiments of the Universities

of Groningen and Leiden.


Plasma and tissue sampling and analysis

The mice were fasted from 6-10 AM. Blood glucose concentrations were measured using a EuroFlash

meter (Lifescan Benelux, Beerse, Belgium). Mice were subsequently sacrifi ced by cardiac puncture

under isofl urane anesthesia. Livers and skeletal muscles were quickly removed, snap-frozen in liquid

nitrogen and stored at -80 °C. Epididymal, perirenal and brown fat pads were weighed. For adipocyte

histology, epididymal fat was fi xed in 4% paraformaldehyde/PBS, and embedded in paraffi n. Blood

was centrifuged (4000 xg for 10 min at 4 °C) and plasma was stored at -20 °C. Plasma TG concentra-

tions were determined using a commercially available kit (Roche Diagnostics, Mannheim, Germany).

Plasma insulin concentrations were determined using ELISA (Ultrasensitive Mouse Insulin kit; Merco-

dia, Uppsala, Sweden). Plasma leptin and adiponectin concentrations were determined by Luminex®

Multiplex technology (Luminex Corporation, Austin, TX) using Multiplex Immunoassays (Mouse adi-

pokine panel; Millipore, Amsterdam, The Netherlands). Hepatic TG content was determined using a

commercial available kit (Roche) after lipid extraction [69]. Hepatic fatty acid content was determined

by gas chromatography after transmethylation [123]. Hepatic glycogen content was determined as

previously described [41]. For adipocyte histology, 3 μm paraffi n sections were stained with hema-

toxylin and eosin and analyzed at 10x magnifi cation. Fat cell area of two representative sections

per group was quantifi ed using image analysis software (Qwin, Leica, Wetzlar, Germany). RNA was

extracted from liver and skeletal muscle using Tri reagent (Sigma-Aldrich, St. Louis, MO, USA). RNA

was converted into cDNA by a reverse transcription procedure using M-MLV (Sigma) and random

primers according to the manufacturer’s protocol. For realtime PCR, cDNA was amplifi ed using the

qPCR core kit (Eurogentec, Seraing, Belgium) and the appropriate primers and probes. Primer and

probe sequences of the following genes have been published (www.LabPediatricsRug.nl): 18S, 36B4,

carnitine-acylcarnitine translocase (Cact), Cd36, Cpt1b/2, carnitine acyltransferase (Crat), G6ph, G6pt,

Pepck and Pgc-1α. The sequences for all other primer and probes are given in Supplemental Table 1.

All mRNA levels were calculated relative to the expression of 18S (liver) or 36B4 (skeletal muscle) and

normalized for expression levels of chow-fed mice.

Indirect calorimetry

We assessed in vivo energy metabolism in high-fat and high-fat/fi sh oil-fed mice using a Compre-

hensive Laboratory Animal Monitoring System (CLAMS; Columbus Instruments, Columbus, USA).

Mice were housed individually to enable real time and continuous monitoring of metabolic gas ex-

change. Detectors measured O2 and CO

2 sequentially across each chamber for 45 seconds at seven-

minute intervals. RER was calculated as the ratio between the volume of CO2 produced (VCO

2) and

the volume of O2 consumed (VO

2). RER values were compared to data obtained in mice receiving

a low-fat diet (Research Diet Services, Wijk bij Duurstede, The Netherlands). Carbohydrate and fat

oxidation rates were calculated from VO2 and VCO

2 using the following formulas [160]:

Carbohydrate oxidation (kcal/h)= ((4.585 x VCO2)-(3.226 x VO

2)) x 4/1000

Fat oxidation (kcal/h)= ((1.695 x VO2)-(1.701 x VCO

2)) x 9/1000

With VO2 and VCO

2 given in mL/h. Total energy expenditure was calculated from the sum of carbo-

hydrate and fat oxidation (i.e., protein oxidation was not accounted for).

7 9Chapter 5

In vivo glucose metabolism

Five days prior to the experiment, mice were equipped with a permanent catheter in the right atri-

um via the jugular vein [72]. The two-way entrance of the catheter was attached to the skull using

acrylic glue. Food was withdrawn nine hours (from 11 PM-8 AM) prior to the start and during the

experiment. The mice were kept in experimental cages and had free access to water. All in vivo infu-

sion experiments lasted six hours and were performed in conscious, unrestrained mice, because the

cages allowed frequent collection of blood spots without the use of anesthesia. During the experi-

ment, blood glucose concentrations were measured every 15 minutes in a small drop of blood that

was taken from the tail vein using a EuroFlash glucose meter. Every 30 minutes, a bloodspot was

collected on fi lter paper via tail bleeding for GC-MS measurements. Basal glucose metabolism was

studied by infusion of [U-13C]-glucose at 7.5 μmol/h for 120 minutes. Subsequently, a hyperinsuline-

mic euglycemic clamp was performed for 240 minutes. During the clamp, mice were infused with

two solutions. The fi rst solution consisted of BSA (1% w/v, Sigma-Aldrich) containing somatostatin

(40 μg/mL, UCB, Breda, The Netherlands), insulin (44 mU/mL, Actrapid; Novo Nordisk, Bagsvaerd,

Denmark), glucose (1078 mM) and [U-13C]-glucose (33 mM, 99% 13C atom %excess; Cambridge Iso-

tope Laboratories, Andover, MA, USA) and was infused at a constant rate of 0.135 mL/h. The second

solution consisted of glucose (1078 mM) and [U-13C]-glucose (33 mM) and its infusion rate was ad-

justed according to the blood glucose concentration in order to maintain euglycemia. Prior to these

experiments, dose-responsiveness of insulin-mediated suppression of hepatic glucose production

and stimulation of glucose clearance was tested in separate groups of mice. We performed hyper-

insulinemic euglycemic clamps using the protocol described earlier and applied diff erent insulin

doses (0-30 mU/hr) and determined at which insulin dose the half-maximal eff ect on peripheral

glucose clearance and hepatic glucose production was reached (Figure 1A/B). This dose (6 mU/hr)

was used for the clamps performed on the animals fed the diff erent diets. At the end of all in vivo

infusion experiments, the mice were sacrifi ced under isofl urane anesthesia.

0 1.5 3 6 15 300

50

100

150

Insulin dose (mU/h)

Hyp

erin

suli

nem

ic h

epat

ic g

luco

se p

rodu

ctio

n ra

te(μ

mol

/kg/

min

)

0 1.5 3 6 15 300

50

100

150

Insulin dose (mU/h)

Hyp

erin

suli

nem

ic m

etab

olic

cle

aran

ce ra

te(m

L/kg

/min

)

A B

Figure 1. Dose-dependent eff ect of insulin on glucose disposal and glucose production under hyperinsulinemic euglycemic clamp

conditions in chow-fed C57Bl/6 mice.

A, Metabolic clearance rates and B, Hepatic glucose production rates. Values represent means ± SEM for n=4-6.


Analysis of in vivo glucose metabolism

Analytical procedures for extraction of glucose from blood spots, derivatization of the extracted

compounds and GC-MS measurements of derivatives were performed according to van Dijk et al.

[40]. Calculations were performed according to Grefhorst et al. [60]. Mean hepatic glucose produc-

tion rates and metabolic clearance rates (a measure of glucose disposal) were calculated for the

period of steady-state isotope dilution.

Downstream insulin receptor signalling

Mice were fasted from 6-10 AM. They subsequently received an intravenous injection of insulin (Ac-

trapid, 7.5 mU dissolved in PBS/BSA 1%, n=3/diet group) or vehicle (PBS/BSA 1%, n=3/diet group).

After 10 min, mice were sacrifi ced by cardiac puncture under isofl urane anesthesia. Livers, skeletal

muscles and epididymal adipose tissue were quickly removed. Homogenates were prepared in lysis-

buff er containing 50 mM Tris/HCl pH 7.5, 150 mM NaCl, 5 mM EDTA, 30 mM sodium pyrophosphate,

50 mM NaF, 1% triton X-100, 1mM phenylmethylsulfonyl fl uoride, 1% phosphatase inhibitor cocktails

I and II (Sigma-Aldrich), and 1 Complete protease inhibitor cocktail tablet (Roche) per 50 mL [45]. For

Western Blotting, samples were electrophoresed using polyacrylamide gels. Proteins were blotted

onto a nitrocellulose fi lter (GE Healthcare, Little Chalfont, UK) by tank blotting. Ponceau S staining

was performed to check for equal protein transfer. Filters were blocked in Tris-buff ered saline (pH 7.4)

containing 0.1% Tween 20 and 4 % skim milk powder. Blots were incubated with primary antibodies

against phospho-Akt ser473, total-Akt (Cell Signaling, Danvers, MA) and Cpt1b (Alpha Diagnostic

International, San Antonio, TX). After washing, immunecomplexes were detected using horseradish

peroxidase conjugated donkey anti-rabbit IgG and Supersignal west pico chemiluminescent sub-

strate (Thermo Scientifi c, Etten-Leur, The Netherlands). Band-densities were determined by using a

Gel Doc XR (Biorad, Hercules CA, USA). PI3K activity was determined as previously described [161].

The incorporated radioactivity was quantifi ed using a phosphorimager.

Statistics

All data are presented as means ± SEM. Statistical analysis was performed using SPSS for Windows

software (SPSS 12.02, Chicago, IL, USA). Analysis of two groups (chow versus high-fat, high-fat versus

high-fat/fi sh oil or insulin versus vehicle) was assessed by Kruskal Wallis/Mann-Whitney U-test for

biometric, plasma and tissue parameters or by ANOVA for repeated measurements for the infusion

experiments. Statistical signifi cance was reached at a p value below 0.05.

8 1Chapter 5

RESULTSFish oil induces additional weight gain and increases plasma adipokine levels in mice fed a high-fat

diet

High-fat and high-fat/fi sh oil feeding resulted in an increased caloric intake compared to chow feed-

ing (Table 1). Mice receiving the fi sh-oil enriched diet gained more weight than those receiving the

high-fat diet. Fish-oil fed animals exhibited the largest increase in adipose tissue mass. Plasma leptin

and adiponectin concentrations were furthermore elevated compared to chow- and high fat-fed

mice (Table 1). Fish oil feeding also resulted in an additional increase in adipocyte size (Figure 2A

and B).

The fat-to-carbohydrate oxidation is increased in mice fed a high-fat or high-fat/fi sh oil diet

Hepatic TG and total fatty acid contents were increased in mice fed the high-fat diet and normalized

in mice fed the high-fat/fi sh oil diet to levels observed in mice fed chow (Table 1). Plasma TG con-

centrations were furthermore decreased in mice fed the high-fat/fi sh oil diet (Table 1). Blood glucose

and plasma insulin concentrations were elevated in high-fat and high-fat/fi sh oil-fed animals (Table

1).

Table 1. Metabolic parameters and metabolite levels in C57Bl/6 mice fed chow, high-fat or high-fat/fi sh oil diets for 6 weeks.

chow high-fat high-fat/fi sh oil

Caloric intake (kcal/24 h) 12.3±0.1 17.1±0.8* 16.8±1.0

Body weight gain (%) 11±2 17±2 27±4#

Epididymal adipose tissue (mg) 393±36 1346±251* 1913±394#

Perirenal adipose tissue (mg) 108±32 406±87* 648±80#

Brown adipose tissue (mg) 72±36 142±24* 193±31#

Blood glucose (mM) 8.7±0.9 9.5±0.6 12.9±0.8#

Plasma insulin (ng/mL) 0.2±0.0 0.7±0.2* 0.8±0.2

Plasma leptin (ng/mL) 1.4±0.2 2.9±0.3 6.3±1.1#

Plasma adiponectin (μg/mL) 13.7±1.0 14.2±1.5 20.7±2.2#

Plasma TG (mM) 0.6±0.1 0.6±0.1 0.4±0.1#

Values represent means ± SEM for n=5-7; * p<0.05 high-fat vs. chow; # p<0.05 high-fat/fi sh oil vs. high-fat (Mann-Whitney U-test).

Mice fed a low-fat diet exhibited a high RER during the dark phase, which decreased during the light

phase (dark: 0.95±0.01, light: 0.90±0.02, p<0.05 dark versus light, Figure 3A), indicating a switch from

whole-body carbohydrate (dark: 0.44±0.02, light: 0.33±0.03 kcal/h, p<0.05 dark versus light) to fat

(dark: 0.08±0.02, light: 0.12±0.02 kcal/h, p<0.05 dark versus light) oxidation. High fat-fed mice exhibi-

ted a low RER during both the dark and light phase (dark: 0.79±0.01, light: 0.79±0.01, Figure 3A). This

reduction in RER was even more pronounced in mice fed the high-fat/fi sh oil diet (dark: 0.75±0.01,

light: 0.76±0.01, Figure 3A). We observed no signifi cant diff erences in RER between dark and light

phase in both high-fat and high-fat/fi sh oil-fed animals (high-fat: p=0.17, high-fat/fi sh oil: p=0.18).


Direct comparison of both high fat-fed groups revealed a signifi cant reduction in 24-h RER in high-

fat/fi sh oil-fed mice (high-fat: 0.79±0.01, high-fat/fi sh oil: 0.75±0.01, p<0.05), indicating an increased

fat-to-carbohydrate oxidation ratio. Absolute 24-h carbohydrate oxidation rates were signifi cantly

lower in animals on high-fat/fi sh oil compared to mice fed high-fat (high-fat: 0.12±0.01, high-fat/fi sh

oil: 0.07±0.01 kcal/h, p<0.05) while absolute 24-h fat oxidation rates only tended to be increased in

these mice (high-fat: 0.28±0.01, high-fat/fi sh oil: 0.32±0.02 kcal/h, p=0.09). Calculated 24-h whole-

body energy expenditure was not diff erent between mice fed the high-fat and high-fat/fi sh oil diets

(high-fat: 0.40±0.01, high-fat/fi sh oil: 0.39±0.01 kcal/h, p=0.15) because of the increased contribution

of fatty acid oxidation to total energy expenditure in high-fat/fi sh oil-fed animals.

The expression of genes involved in fatty acid uptake and oxidation was elevated in skeletal mus-

cle of mice fed the high-fat/fi sh oil diet compared to mice fed chow and high-fat diet (Figure 3B).

Fish oil furthermore decreased Pgc-1α mRNA expression (Figure 3B) and both high-fat diets slightly

increased Cpt1b protein expression (Figure 3C) in muscles.

0-2 2-4 4-6 6-8 8-10 10-12 12-14 14-160

10

20

30

40

50

60

70 chowhigh-fathigh-fat/fish oil

Adipocyte area (*1000 μm2)

Adip

ocyt

e siz

e di

strib

utio

n(%

of

tota

l num

ber)

Figure 2. Adipocyte morphology in C57Bl/6 mice fed chow, high-fat

and high-fat/fi sh oil diets for 6 weeks.

A, Adipocyte morphology and B, Adipocyte size distribution in repre-

sentative samples of epididymal fat tissue.

Open bars, chow diet; fi lled bars, high-fat diet; dotted bars, high-fat/

fi sh oil diet.

B

A


8 3Chapter 5

Fish oil aggravates impaired basal and insulin-stimulated glucose clearance in high fat-fed mice

We determined basal and insulin-stimulated glucose clearance in mice receiving chow, high-fat and

high-fat/fi sh oil diets. Basal glucose clearance rates were slightly reduced in high fat-fed mice (chow:

19±2, high-fat: 16±1 mL/kg/min, p=0.09) but signifi cantly impaired in mice fed the high-fat/fi sh oil

diet (high-fat/fi sh oil: 12±1 mL/kg/min, p<0.05 high-fat/fi sh oil versus high-fat, Figure 4). Continuous

insulin infusion (6 mU/h) increased glucose clearance rates in all groups. The hyperinsulinemic glu-

cose clearance rates (chow: 71±2, high-fat: 54±2, high-fat/fi sh oil: 31±3 mL/kg/min, p<0.05 high-fat

versus chow and high-fat/fi sh oil versus high-fat, Figure 4) and the insulin-mediated stimulation of

glucose clearance were clearly impaired in high fat-fed mice. Insulin action was even further dete-

riorated by fi sh oil (p<0.05 high-fat versus chow and high-fat/fi sh oil versus high-fat, Figure 4). Blood

glucose concentrations and the glucose infusion rates that were required to maintain euglycemia

during the hyperinsulinemic clamps are given in Table 2. Glucose infusion rates were lower in mice

receiving the high-fat diet and further reduced by fi sh oil.

Table 2. Blood glucose concentrations and glucose infusion rates required to maintain euglycemia under hyperinsulinemic conditions

in C57Bl/6 mice fed chow, high-fat or high-fat/fi sh oil diets for 6 weeks.


Blood glucose concentration (mM) 7.0±0.2 7.1±0.2 6.9±0.2

Glucose infusion rate (μmol/kg/min) 443±14 304±14* 145±10#

Values represent means ± SEM for n=5-9 during stable isotope infusion (t=270-360 min); * p<0.05 high-fat vs. chow; # p<0.05 high-fat/

fi sh oil vs. high-fat (ANOVA).

Figure 3. Substrate utilization in C57Bl/6 mice fed a low-fat

diet, high-fat and high-fat/fi sh oil diets for 6 weeks.

A, RER. Dashed line, low-fat diet; black line, high-fat diet; grey

line, high-fat/fi sh oil diet. B, Expression of genes involved in

fat oxidation and C, Cpt1b protein expression.

Open bars, chow diet; fi lled bars, high-fat diet; dotted bars,

high-fat/fi sh oil diet. Average RER values represent means

for n=7-8. Average qPCR values represent means ± SEM for

n=5-7. Average Western Blot values represent means ± SEM

for n=3; * p<0.05 high-fat vs. chow; # p<0.05 high-fat/fi sh oil

vs. high-fat (Mann-Whitney U-test).

0.7

0.8

0.9

1.0

1.1 high-fathigh-fat/fish oil

low-fat

dark light dark light dark

Res

pira

tory

Exc

hang

e R

atio

Pgc-1α

Cd36 AcsCpt1b

Cpt2 Crat Cact Lcad

0

1

2

chowhigh-fathigh-fat/fish oil

Rela

tive

mRN

A ex

pres

sion

# ##

A B

0.0

0.5

1.0

1.5


Cpt1

b pr

otei

n ex

pres

sion

(A.

U.)

*

C


Fish oil suppresses basal and insulin-stimulated hepatic glucose production in high fat-fed mice

The hepatic expression of the gluconeogenic genes Pgc1α, Pepck, G6ph and G6pt was decreased in

mice fed the high-fat/fi sh oil diet (Figure 5A). Hepatic glycogen content was reduced in these ani-

mals (chow: 334±24, high-fat: 357±37, high-fat/fi sh oil: 269±26 μmol/g p<0.05 high-fat/fi sh oil ver-

sus high-fat). High-fat feeding did not aff ect basal hepatic glucose production rates (chow: 128±6,

high-fat: 136±5 μmol/kg/min, p=0.42, Figure 5B) while these were reduced in high-fat/fi sh oil fed

mice compared to mice fed chow or high-fat (high-fat/fi sh oil: 102±5 μmol/kg/min, p<0.05 high-fat/

fi sh oil versus high-fat, Figure 5B). Continuous insulin infusion (6 mU/h) suppressed hepatic glucose

production rates in all mice. Clamped hepatic glucose production rates were higher in high fat-

fed mice compared to chow-fed mice, (chow: 43±10, high-fat: 75±8 μmol/kg/min, p<0.05 high-fat

versus chow, Figure 5B) and the insulin-mediated suppression of hepatic glucose production rates

was blunted in these animals (p<0.05 high-fat versus chow, Figure 5B). Fish oil partially normalized

clamped hepatic glucose production rates to control values (high-fat/fi sh oil: 55±7 μmol/kg/min,

p=0.15 high-fat/fi sh oil versus high-fat, Figure 5B). The insulin-mediated suppression of hepatic glu-

cose production in these animals was similar to high-fat-fed mice (p=0.70, Figure 5B).

chow high-fat high-fat/fish oil0

50

100

basalclamp

Met

abol

ic c

lear

ance

rate

(mL/

kg/m

in)

*

#

+292±24% +253±22% +168±22%

Figure 4. Peripheral glucose clearance in C57Bl/6 mice fed chow, high-fat and high-fat/fi sh oil diets for 6 weeks.

Open bars, basal period; dotted bars, hyperinsulinemic clamp. Inset, relative increase of metabolic clearance rates from basal to hyper-

insulinemic period. Values represent means ± SEM for n=5-9 during steady state infusion (t=75-120 min for basal period; t=270-360 min

for hyperinsulinemic clamp); # p<0.05 high-fat/fi sh oil vs. high-fat (ANOVA).

B


50

100

150

200basalclamp

Hep

atic

glu

cose

pro

duct

ion

(μm

ol/k

g/m

in)

*

$

-66±7% -44±7% -45±7%

Pgc-1α

Pepck G6ptG6ph

0

1

2chowhigh-fathigh-fat/fish oil

Rela

tive

mRN

A ex

pres

sion

#

# #

*

*

Figure 5. Hepatic glucose production in C57Bl/6 mice fed chow, high-fat and high-fat/fi sh oil diets for 6 weeks.

A, Expression of genes involved in glucose production.

Open bars, chow diet; fi lled bars, high-fat diet; dotted bars, high-fat/fi sh oil diet. B, Hepatic glucose production. Open bars, basal period;

dotted bars, hyperinsulinemic clamp. Inset, relative decrease of hepatic glucose production from basal to hyperinsulinemic period. Va-

lues represent means ± SEM for n=5-9 during steady state infusion (t=75-120 min for basal period; t=270-360 min for hyperinsulinemic

clamp); * p<0.05 high-fat vs. chow; # p<0.05 high-fat/fi sh oil vs. high-fat (Figure A, Mann-Whitney U-test; Figure B, ANOVA).

A

8 5Chapter 5

Insulin-stimulated PI3K activity is attenuated in adipose tissue and livers of mice fed a high-fat or

high-fat/fi sh oil diet

Insulin signifi cantly increased IRS1-associated PI3K activity in the muscles of all mice (Figure 6A).

Basal IRS1/PI3K activity was non-signifi cantly increased in the adipose tissue of high-fat and high-fat/

fi sh oil-fed animals, while insulin’s stimulatory eff ect on this activity was completely abolished (Fig-

ure 6B). Furthermore, high-fat and high-fat/fi sh oil feeding impaired hepatic insulin responsiveness

of both IRS1 (Figure 6C) and IRS2-associated PI3K activity (Figure 6D). The changes in PI3K activity

were however not refl ected in the expression of phosphorylated Akt (Table 3).


100

200

300

400

500

vehicleinsulin

IRS1

-ass

ocia

ted

PI3K

act

ivity

(A.U

.)

*

*

*+132%

+239%

+102%


100

200

300 vehicleinsulin

IRS1

-ass

ocia

ted

PI3K

act

ivity

(A.U

.)

*+68%

*+22%

*+22%


100

200

300

vehicleinsulin

IRS1

-ass

ocia

ted

PI3K

act

ivity

(A.U

.)

*+83% +13%

-17%


100

200

300 vehicleinsulin

IRS2

-ass

ocia

ted

PI3K

act

ivity

(A.U

.)

*+50%

+34% +34%

A B

C D

Figure 6. IRS-associated PI3K activity in C57Bl/6 mice fed chow, high-fat and high-fat/fi sh oil diets for 6 weeks.

A, IRS1-associated PI3K activity in skeletal muscle; B IRS1-associated PI3K activity in adipose tissue; C, IRS1-associated PI3K activity in liver

and D, IRS2-associated PI3K activity in liver.

Open bars, vehicle; dotted bars, insulin. Inset, relative increase by insulin compared to vehicle. Values represent means ± SEM for n=3. *

p<0.05 insulin vs. vehicle (Mann-Whitney U-test).


Table 3. Akt phosphorylation in muscle, adipose tissue and liver of C57Bl/6 mice fed chow, high-fat or high-fat /fi sh oil diets for 6

weeks.


vehicle insulin vehicle insulin vehicle insulin

Muscle

pAkt ser473 1.0±0.1 12.5±0.9* 1.1±0.1 11.5±0.9* 1.1±0.2 13.1±0.4*

tAkt 1.0±0.1 0.7±0.1 1.0±0.1 0.7±0.1* 1.0±0.0 0.8±0.1

ratio pAkt/tAkt 1.0 17.8 1.1 16.4 1.1 16.8

Adipose tissue

pAkt ser473 1.0±0.2 9.7±0.4* 1.6±0.4 15.4±2.9* 1.3±0.3 13.8±2.9*

tAkt 1.0±0.2 0.7±0.0 1.3±0.2 1.0±0.1 1.4±0.0 1.0±0.1*

ratio pAkt/tAkt 1.0 13.1 1.2 15.1 0.9 14.1

Liver

pAkt ser473 1.0±0.1 11.4±3.1* 1.1±0.2 7.3±3.4* 1.0±0.2 9.4±1.1*

tAkt 1.0±0.1 0.9±0.1 1.0±0.1 0.9±0.1 0.8±0.1 0.8±0.0

ratio pAkt/tAkt 1.0 12.7 1.1 7.9 1.2 12.5

Values represent means ± SEM for n=3. * p<0.05 insulin vs. vehicle (Mann-Whitney U-test).

8 7Chapter 5

DISCUSSIONHigh-fat diets predispose to development of insulin resistance and type 2 diabetes [145]. Studies

on the eff ect of dietary fi sh oil supplementation on glucose control are inconclusive [144–150]. n-3

PUFA from fi sh oil are known to suppress hepatic fatty acid synthesis, while they simulate fatty acid

oxidation. The relationship between fatty acid oxidation and glucose disposal is however unclear

[74,153–159] and in vivo data on the concurrent alterations in substrate utilization and glucose me-

tabolism upon fi sh oil substitution of a high-fat diet are lacking. This prompted us to study the meta-

bolic eff ects of fi sh oil in mice fed a high-fat diet. C57Bl/6 mice were subjected to a 6-week dietary

challenge by a diet rich in beef tallow or a similar diet in which part of the tallow was replaced by fi sh

oil and outcome was compared to animals that were fed normal laboratory chow.

Mice fed the high-fat/fi sh oil diet exhibited a larger body weight gain compared to mice fed the

high-fat diet and histological analysis revealed remarkable adipocyte enlargement in these animals.

Accordingly, plasma levels of leptin were increased in high-fat/fi sh oil-fed mice. A similar phenotype

in high fat-fed mice receiving fi sh oil has recently been published by Coenen et al. [162]. Other

studies have, however, reported a body weight-reducing eff ect of fi sh oil in obese diabetic mice

[163–165], in mice on high α-linoleic acid diets [166] and during long-term dietary intervention

[167,168]. Altogether, these data suggest that the metabolic state interferes with the eff ects of fi sh

oil on adipose tissue development [169].

High-fat feeding clearly altered the circadian pattern of substrate utilization. Animals receiving

chow diet switched from glucose oxidation during the dark phase to fat oxidation during the light

phase. These day-night variations in substrate utilization correlate with those reported for plasma

insulin concentrations [170]. Kohsaka et al. [171] observed disturbed circadian rhythmicity of serum

insulin concentrations (i.e., increased insulin concentrations during both day and light phases) upon

high-fat feeding in mice. In the current study, mice fed a high-fat diet did not switch from fat oxida-

tion during the light phase toward carbohydrate oxidation during the dark phase. These animals

displayed sustained high fat-to-carbohydrate oxidation ratios and, thus, insulin responsiveness of

substrate utilization must have been disturbed. RER values in mice fed the high-fat/fi sh oil diet were

consistently lower than those observed in animals receiving high-fat and the fat-to-carbohydrate

oxidation ratio was further increased. Energy expenditure was similar in high-fat and high-fat/fi sh oil-

fed mice and the increased fat oxidation therefore coincided with a reduction in glucose oxidation.

Impaired ability to switch between glucose and lipid oxidation has been reported in obese and/

or diabetic subjects [172,173] and metabolic infl exibility towards glucose utilization is related to

impaired glucose clearance in type 2 diabetic subjects [174]. The current study provides a detailed

analysis on the consequences of a sustained reliance on fat oxidation in high-fat and high-fat/fi sh

oil-fed mice. Basal glucose clearance was reduced in high-fat and high-fat/fi sh oil-fed animals com-

pared to mice receiving chow. Insulin-mediated stimulation of glucose clearance was impaired by

high-fat feeding, indicative for peripheral insulin resistance. This eff ect was clearly potentiated by

fi sh oil. The reduced rates of carbohydrate oxidation during the light and dark phase in high-fat

and high-fat/fi sh oil-fed mice were thus paralleled by a further reduction in glucose clearance. This

phenotype, which is characterized by an increased whole body fat-to-carbohydrate oxidation ratio

and a concomitant deterioration of insulin sensitivity, has also been observed in type 2 diabetic

subjects receiving long-term fi sh oil supplementation [146]. Skeletal muscle is the major site for


insulin-stimulated glucose clearance [175] and an enhanced fatty acid oxidation in skeletal muscle

is associated with impaired glucose metabolism [153–156]. An inhibition of fatty acid uptake and

oxidation, on the other hand, enhances glucose disposal in vivo in normal and insulin-resistant mice

[74,153,158,159]. It is therefore tempting to speculate that the peripheral insulin resistance in high-

fat and high-fat/fi sh oil-fed mice mainly resulted from an impaired glucose disposal into muscles

[174]. Our studies do, however, not provide evidence for an impairment in downstream insulin re-

ceptor signalling upon high-fat and high-fat/fi sh oil feeding. We did not observe a reduction in IRS1-

associated PI3K activity in skeletal muscles of high-fat-fed mice, while insulin’s stimulatory action on

PI3K activity was to some extent attenuated by high-fat/fi sh-oil feeding. The obese phenotypes of

the high-fat and high-fat/fi sh oil-fed mice presumably contributed to the impairment of glucose

disposal, because the insulin-dependent increase in PI3K activity in adipose tissues was blunted in

both groups. The increase in plasma adiponectin concentrations upon fi sh oil replacement was not

suffi cient to counterbalance the insulin resistance in high fat-fed mice.

Basal hepatic glucose production rates were comparable in chow and high fat-fed mice (Figure

5B), despite the fasting hyperinsulinemia that was observed in animals receiving high-fat. Consistent

with this hepatic insulin insensitivity under basal conditions, clamped hepatic glucose production

rates were higher in high fat-fed mice as compared to animals receiving chow. Basal hepatic glucose

production was lower in high-fat/fi sh oil-fed mice (-25% vs. high-fat), suggestive for an improved he-

patic insulin responsiveness under these conditions. Hepatic gluconeogenic gene expression levels

were concomitantly reduced and hepatic glycogen content was lower in high-fat/fi sh oil-fed ani-

mals. These observations suggest a decreased gluconeogenic fl ux in these mice [75,80]. However,

additional in vivo isotope studies are needed to quantify the actual gluconeogenic fl ux in fi sh oil

fed-mice. Fish oil partially normalized hepatic glucose production under clamped conditions (-27%

vs. high-fat). However, as a result of the reduced basal hepatic glucose production upon fi sh oil re-

placement, the relative suppression of hepatic glucose production from basal to hyperinsulinemic

conditions was similarly impaired in high-fat and high-fat/fi sh-oil fed mice. This is supported by the

observation that insulin-stimulated PI3K activity was lower in the livers of animals receiving either of

the two high-fat diets. The lower basal glucose production in high-fat/fi sh oil fed mice may therefore

not just result from a restoration of hepatic insulin sensitivity. As yet unidentifi ed mechanisms may

contribute to the lowering of basal hepatic glucose production upon fi sh oil replacement.

Interestingly, Neschen et al. [176] also performed a study on fi sh oil replacement of a vegetable

oil-rich diet in mice on a SV129 background. Compared to the current study, these authors observed

major diff erences in the response to the dietary challenges. Strikingly, Neschen et al. did not report

peripheral insulin resistance in mice either fed the saffl ower oil or the saffl ower/fi sh oil diet for 2

weeks. These authors observed severe hepatic insulin resistance in saffl ower oil-fed animals, which

was partially prevented by fi sh oil. Besides diff erences in duration of dietary intervention, fat sources

and the genetic background of the mice [177], the experimental conditions under which glucose

metabolism was studied were diff erent from ours. This may be of importance [178]. We performed

6-h stable isotope infusion studies in awake, freely-moving mice. In this experimental setup, insu-

lin dose-dependently increases glucose clearance while it suppresses hepatic glucose production

(Figure 1). Glucose demand by organs and tissues is diminished if animals are anaesthetized or re-

strained (unpublished observations). Therefore, in our opinion, most relevant experimental data are

8 9Chapter 5

obtained from studies in which normal physiology is minimally disturbed.

In summary, we have shown that fi sh oil alters substrate utilization by increasing the fat-to-car-

bohydrate oxidation ratio. This is associated with a further detoriation of insulin-mediated glucose

clearance in mice fed a high-fat diet. Our data indicate that an increased fat-to-carbohydrate oxi-

dation ratio per se does not prevent adiposity and impaired glucose clearance, and emphasizes the

need for a change in energy balance to arrest diet-induced obesity and peripheral insulin resistance.

These insights will allow us to defi ne the metabolic conditions under which dietary approaches may

be useful to prevent insulin resistance and type 2 diabetes.

ACKNOWLEDGEMENTSThe authors thank Vincent W. Bloks for scientifi c discussion and Juul F.W. Baller, Trijnie Bos and Theo

Boer for excellent technical assistance.

This work was supported by the Nutrigenomics Consortium (NGC) and the Center of Medical Sys-

tems Biology (CMSB), established by the Netherlands Genomics Initiative/Netherlands Organization

for Scientifi c Research (NGI/NWO).

6 High-fat feeding induces hepaticfatty acid elongation in mice

M.H. Oosterveer

T.H. van Dijk

U.J.F. Tietge

T. Boer

R. Havinga

F. Stellaard

A.K. Groen

F. Kuipers

D-J. Reijngoud

ADAPTED FROMPLOS ONE. 2009 26;4(6):E6066

9 2 Dietary fatty acids alter lipogenic fl uxes

ABSTRACTHigh-fat diets promote hepatic lipid accumulation. Paradoxically, these diets also induce lipogenic

gene expression in rodent liver. Whether high expression of these genes actually results in an in-

creased fl ux through the de novo lipogenic pathway in vivo has not been demonstrated.

To interrogate this apparent paradox, we have quantifi ed de novo lipogenesis in C57Bl/6J mice

fed either chow, a high-fat or a n-3 PUFA-enriched high-fat diet. A novel approach based on MIDA

following 1-13C acetate infusion was applied to simultaneously determine de novo lipogenesis, fatty

acid elongation as well as cholesterol synthesis. Furthermore, we measured VLDL-TG production

rates. High-fat feeding promoted hepatic lipid accumulation and induced the expression of lipo-

genic and cholesterogenic genes compared to chow-fed mice: induction of gene expression was

found to translate into increased oleate synthesis. Interestingly, this higher lipogenic fl ux (+74 μg/

g/h for oleic acid) in mice fed the high-fat diet was mainly due to an increased hepatic elongation

of unlabeled palmitate (+ 66 μg/g/h) rather than to elongation of de novo synthesized palmitate.

In addition, fractional cholesterol synthesis was increased, i.e. 5.8±0.4% vs. 8.1±0.6% for control and

high fat-fed animals, respectively. Hepatic VLDL-TG production was not aff ected by high-fat feeding.

Partial replacement of saturated fat by fi sh oil completely reversed the lipogenic eff ects of high-fat

feeding: hepatic lipogenic and cholesterogenic gene expression levels as well as fatty acid and cho-

lesterol synthesis rates were normalized.

In conclusion, high-fat feeding induces hepatic fatty acid synthesis in mice, by chain elongation

and subsequent desaturation rather than de novo synthesis, while VLDL-TG output remains unaf-

fected. Suppression of lipogenic fl uxes by fi sh oil prevents from high fat diet-induced hepatic stea-

tosis in mice.

9 3Chapter 6

INTRODUCTIONNon-alcoholic fatty liver disease (NAFLD) is one of the hallmarks of the metabolic syndrome and is

strongly associated with obesity and insulin resistance [179]. NAFLD is characterized by the accumu-

lation of hepatic TGs resulting from an imbalance between uptake, synthesis, export and oxidation

of fatty acids [180]. NAFLD may progress to non-alcoholic steatohepatitis (NASH) in response to a

‘second hit’ [181]. Although high fat diets consistently induce hepatic steatosis and insulin resistance

in humans and laboratory animals [182–184], the mechanisms underlying this high fat diet-induced

lipid accumulation are largely unknown.

Interestingly, high-fat feeding has been reported to result in a paradoxical increase in the expres-

sion of lipogenic genes in mouse liver. This was suggested to be mediated via PGC-1ß co-activation

of the lipogenic transcription factor SREBP-1c [185]. Ablation or suppression of critical genes control-

ling hepatic lipogenesis [130,186–188] counteracts the development of hepatic steatosis in animals

receiving high-fat diets. Furthermore, partial substitution of the fat within a high-fat diet for fi sh oil, a

source of n-3 PUFA, abrogates hepatic lipid accumulation [184]. An inhibition of the activity of lipo-

genic transcription factors and the subsequent suppression of their target genes by n-3 PUFA [189]

is considered to contribute to the protective eff ects of fi sh oil. Although these observations suggest

that the activity of lipogenic enzymes is related to the degree of high fat diet-induced hepatic stea-

tosis, an increased de novo fatty acid synthesis appears counterintuitive under conditions of a high

dietary fatty acid load.

Accurate quantifi cation of fatty acid synthesis and its contribution to hepatic lipid content has not

been reported. Furthermore, the relative contributions of de novo lipogenesis (i.e., synthesis from

acetyl-CoA moieties) and chain elongation of fatty acids to hepatic lipid synthesis in vivo in mice

are currently unknown. In addition, the relationships between high-fat feeding, fatty acid synthe-

sis and hepatic VLDL-TG production are of particular interest because of the reported alterations

in plasma (VLDL)-TG levels following Srebp1 and Pgc-1ß overexpression and knockdown in mice

[185,190]. We therefore determined in vivo rates of fatty acid and cholesterol synthesis in relation to

VLDL-TG production rates in mice fed low-fat laboratory chow, a high-fat diet containing beef fat

(rich in saturated fat), or a diet in which part of the beef fat was replaced by fi sh oil. A novel approach

based on 13C-acetate incorporation followed by MIDA enabled us to quantify the relative contribu-

tion of the de novo lipogenic pathway as well as chain elongation of de novo synthesized and pre-

existing palmitate (C16:0) to stearic acid (C18:0) synthesis and its subsequent desaturation to oleic

(C18:1 n-9) acid. We found that high-fat feeding indeed increased oleic acid synthesis, however this

was mainly due to chain elongation of pre-existing palmitate rather than to an increase in de novo

lipogenesis. Cholesterol synthesis was also increased while VLDL-TG secretion remained unaff ected.

These metabolic changes contributed to hepatic TG and CE accumulation. Fish oil reduced both de

novo lipogenesis and chain elongation and normalized hepatic lipid contents.


EXPERIMENTAL PROCEDURESAnimals and experimental design

Male C57Bl/6J mice (Charles River, L’Arbresle Cedex, France), three months of age, were housed in a

light- and temperature-controlled facility (lights on 6:30 AM-6:30 PM, 21 °C). They were divided into

groups and fed three diff erent diets for six weeks. All diets were obtained from Abdiets, Woerden,

The Netherlands. One group received normal laboratory chow (RMH-B), the second group received

high-fat diet (beef tallow, which is rich in saturated fat) and the third group received a diet in which

42% (w/w) of the beef fat was replaced by fi sh oil (menhaden oil). For diet composition see Table 1.

The fi sh oil-containing diet was refreshed three times a week to prevent oxidation. To exclude acute

postprandial eff ects without the induction of a fasting response, mice were subjected to a short fast-

ing period of 4 hours (6-10 AM) prior to all experiments. Experimental procedures were approved by

the Ethics Committee for Animal Experiments of the University of Groningen.

Table 1. Fatty acid composition of experimental diets and livers.


Diet (mg/g)

C14:0 0.5 12.2 16.1

C16:0 8.4 92.5 79.5

C16:1 0.7 11.5 18.0

C18:0 3.7 76.3 50.5

C18:1 13.7 133.2 101.0

C18:2 16.9 11.5 9.7

C18:3 1.9 2.9 15.2

C20-22 0.4 4.0 53.3

C16 desaturation index 0.08 0.12 0.18

C18 desaturation index 3.7 1.7 2.0

Liver (mg/g)

C14:0 0.1±0.0 0.2±0.0* 0.1±0.0#

C16:0 7.7±0.6 9.0±0.3 8.2±0.6

C16:1 0.6±0.1 1.0±0.1* 0.4±0.1#

C18:0 4.1±0.3 4.8±0.2 5.0±0.2

C18:1 6.0±0.4 16.9±2.1* 6.1±0.6#

C18:2 7.0±0.7 2.8±0.2* 2.9±0.3

C18:3 0.3±0.0 0.2±0.0* 0.2±0.0

C20-22 8.4±0.5 9.2±0.5 11.1±0.6#

C16 desaturation index 0.07±0.00 0.11±0.01* 0.05±0.00#

C18 desaturation index 1.5±0.1 3.6±0.5* 1.2±0.1#

Values represent means ± SEM for n=6/7, * p<0.05 high-fat vs. chow; # p<0.05 high-fat/fi sh oil vs. high-fat (Conover test).

9 5Chapter 6

Hepatic lipid content and gene expression levels

Mice were sacrifi ced by cardiac puncture under isofl urane anaesthesia. Epididymal, perirenal and

brown adipose fat pads were removed and weighed. Livers were quickly removed, weighed, freeze-

clamped and stored at -80 °C. Hepatic TG, total and free cholesterol content were analyzed using

commercial available kits (Roche Diagnostics, Mannheim, Germany and Wako Chemicals, Neuss,

Germany) after lipid extraction [69]. CE contents were calculated as the diff erence between total

and free cholesterol. Hepatic phospholipid content was determined as described previously [191].

Hepatic fatty acid composition was analyzed by gas chromatography after transmethylation using

C17:0 as internal standard [123]. C16 and C18 desaturation indices were calculated from the ratios

between C16:1 n-7 and C16:0 and C18:1 n-7/n-9 and C18:0, respectively.

RNA was extracted from livers using Tri reagent (Sigma-Aldrich, St. Louis, MO) and converted into

cDNA by a reverse transcription procedure using M-MLV and random primers according to the

manufacturer’s protocol (Sigma-Aldrich). For quantitative qPCR, cDNA was amplifi ed using the ap-

propriate primers and probes. Primer and probe sequences for 18S, Acc1 and -2, Acat-1 and -2, Dgat1

and 2, Gpat, Fas, 3-hydroxy-3-methylglutaryl-CoA reductase (Hmgr), Scd1, Srebp-1c and -2 have been

published (www.LabPediatricsRug.nl). For other primer and probe sequences, see Supplemental Ta-

ble 3. mRNA levels were calculated relative to 18S expression and normalized for expression levels

of control mice on chow.

Determination of de novo lipogenesis, chain elongation and cholesterol synthesis in vivo

Mice were equipped with a permanent jugular vein catheter [72] and were allowed a recovery pe-

riod of 4-5 days. All infusion experiments were performed in conscious, unrestrained mice. A 0.3

mol/L sodium [1-13C]-acetate (99 atom %, Isotec/Sigma-Aldrich) solution was infused at a rate of

0.6 mL/hr during 6 hours. Every hour a blood sample was taken via tail bleeding on fi lter paper to

determine fractional cholesterol synthesis rates. At the end of the infusion period, animals were

sacrifi ced by cardiac puncture. Livers were quickly removed, freeze-clamped and stored at -80 °C.

Liver homogenates were prepared in PBS and C17:0 was added as internal standard. Lipids were hy-

drolyzed in HCl/acetonitrile (1:22 v/v) for 45 minutes at 100 °C. Fatty acids were extracted in hexane

and derivatized for 15 minutes at room temperature using α-Br-2,3,4,5,6-pentafl uorobenzyl (PFB)/

acetonitrile/triethanolamine (1:6:2 v/v). Derivatization was stopped by adding HCl and fatty acid-PFB

derivatives were extracted in hexane. Total cholesterol was extracted from blood spots using etha-

nol/acetone (1:1 v/v). Unesterifi ed cholesterol from blood spots was subsequently derivatized using

N,O-bis-(trimethyl)trifl uoroacetamide with 1 % trimethylchlorosilane at room temperature.

The fatty acid-PFB isotopomer patterns were analyzed using an Agilent 5975 series GC/MSD

(Agilent Technologies, Santa Clara, CA). Gas chromatography was performed using a ZB-1 column

(Phenomenex, Torrance, CA). Mass spectrometry analysis was performed by electron capture nega-

tive ionization using methane as moderating gas. Cholesterol-TMS isotopomer patterns were ana-

lyzed using a Trace MS plus GC-MS (Interscience, Breda, The Netherlands). Gas chromatography was

performed using a DB-17 column (J&W Scientifi c, Falson, CA). Mass spectrometry analysis was per-

formed in the electron impact mode.

The normalized mass isotopomer distributions measured by GC-MS (m0-m

x) were corrected for

natural abundance of 13C by multiple linear regression as described by Lee et al. [39] to obtain the ex-

cess fractional distribution of mass isotopomers (M0-M

x) due to incorporation of the infused labeled


compound, i.e., [1-13C]-acetate. This distribution was used in MIDA algorithms to calculate isotope

incorporation and dilution according to Hellerstein et al. [36–38] in order to determine fractional

palmitate synthesis rates. In short, incorporation of [1-13C]-acetate into palmitate was assumed to

solely result from de novo lipogenesis via the malonyl-CoA/FAS pathway. The measured M1 and M

3

isotopomers of palmitate were used to calculate the acetyl-CoA precursor pool enrichment (pacetate

)

and fractional palmitate synthesis (fC16:0

).

Stearate is synthesized by chain elongation of de novo synthesized and/or pre-existing palmitate.

The M1 mass isotopomer of stearate represents the sum of these two processes, while the M

3 mass

isotopomer solely results from chain elongation of labeled palmitate. The following approach was

used to calculate fractional stearate and oleate synthesis. We assumed that the acetate enrichment

used for elongation of palmitate equals pacetate. Stearate generated from de novo synthesized

palmitate was consequently considered as a nonamer of acetate. Therefore, we applied MIDA algo-

rithms using M3(stearate) and pacetate to calculate fractional stearate synthesis from elongation of

de novo synthesized palmitate (fstearate(palmitateDNL

)). Total M1(stearate) was subsequently correct-

ed for the contribution of single labeled stearate originating from elongation of de novo synthesized

palmitate M1(stearate(palmitate

DNL)) to obtain the contribution of single labeled stearate originating

from elongation of pre-existing palmitate M1(stearate(palmitate

PE)). Since we assumed that pacetate

represents the precursor pool enrichment of acetate used in elongation of pre-existing palmitate,

the contribution of elongation of pre-existing palmitate to stearate synthesis fstearate(palmitatePE)

could

fi nally be calculated.

fstearate(palmitateDNL)

= M3(stearate) / F

3(stearate)

in which F3(stearate) equals the theoretical undiluted frequency of triple labeled stearate at p

acetate.

F3(stearate) =

(9)!(9-3)!(3)!

(pacetate

)3(1-pacetate

)9-3

M1(stearate(palmitate

DNL)) is calculated according to:


DNL)) = f

stearate(palmitateDNL) * F

1(stearate)

in which F1(stearate) equals the theoretical undiluted frequency of single labeled stearate at p

acetate.

F1(stearate) =

(9)!(9-1)!(1)!

(pacetate

)1(1-pacetate

)9-1

Consequently:


PE))= M

1(stearate) - M

1(stearate(palmitate

DNL))

in which M1(stearate) represents the measured total M

1 mass isotopomer in stearate.

And fi nally:

fstearate(palmitate)

= M1(stearate(palmitate

PE)) / p

acetate

Oleate is synthesized by desaturation of stearate via SCD1 activity. We used the measured M1 and

M3 isotopomers of oleate to calculate the fractional contributions of chain elongation of de novo

synthesized and pre-existing palmitate to stearate as a direct precursor for oleate (foleate(palmitateDNL)

and

foleate(palmitatePE)

, respectively) using similar equations to that of stearate.

9 7Chapter 6

foleate(palmitateDNL)

= M3(oleate) / F

3(oleate)

in which F3(oleate) equals the theoretical undiluted frequency of triple labeled oleate, calculated as

for stearate using pacetate

.

M1(oleate(palmitate

DNL)) = f

oleate(palmitateDNL) * F

1(oleate)

in which M1(oleate(palmitate

DNL)) represents the contribution of elongation of de novo synthesized

palmitate to M1(oleate) and F

1(oleate) equals the theoretical undiluted frequency of single labeled

oleate, calculated as for stearate using pacetate

.

M1(oleate(palmitate

PE))= M

1(oleate) - M

1(oleate(palmitate

DNL))

in which M1(oleate(palmitate

PE)) represents the contribution of elongation of pre-existing palmitate

to M1(oleate).

foleate(palmitateDNL)

= M1(oleate(palmitate

PE)) / p

acetate

Fractional cholesterol synthesis was calculated on regular time points during isotope infusion (ft)

by MIDA. From this, fractional cholesterol synthesis at infi nite time (f∞) was calculated using SAAM II

software (version 1.2.1 Saam Institute, University of Washington) and the following formula:

ft= f

∞ (1-e-kt)

in which k represents the rate constant.

In vivo VLDL-TG production

Mice were injected intraperitoneally with Poloxamer 407 (1 g/kg body weight) as a 50 mg/mL solu-

tion in saline as previously described [192]. Blood samples were drawn by retro-orbital bleeding into

heparinized tubes at 0, 30, 120, and 240 min after injection. Immediately after the last blood draw,

animals were sacrifi ced by cardiac puncture under isofl urane anaesthesia. Blood was centrifuged (10

minutes, 4000xg) to obtain plasma. Plasma TG levels and TG production rates were determined as

described [192]. Nascent VLDL (d<1.006) was isolated from the fi nal plasma sample of each animal

using a Optima TM LX tabletop ultracentrifuge (Beckman Instruments Inc., Palo Alto, CA) at 108,000

rpm for 125 minutes.

VLDL composition and particle size

VLDL-TG and cholesterol contents were determined as described [192]. Phospholipid content was

determined using a commercial kit (Wako Chemicals). VLDL particle diameter was estimated ac-

cording to Fraser et al. [193]. VLDL particle volume was subsequently derived from its diameter.

Apolipoprotein B (apoB) content of nascent VLDL particles was determined using Western Blot as

previously described [51]. Four representative VLDL samples per group were analyzed and equal

amounts of total lipid were loaded onto the gel. Signal intensity was quantifi ed using a Molecular

Imager (ChemiDoc XRS System, Bio-Rad Laboratories, Hercules, CA) and the relative abundance of

apoB48 versus apoB100-associated particles was calculated.


Statistics

All data are presented as means ± SEM. Statistical analysis was performed using Brightstat software

(www.brightstat.com). Analysis of two groups (chow vs. high-fat, high-fat vs. high-fat/fi sh oil) was

assessed by Kruskal-Wallis using the Conover test for post-hoc analysis. Statistical signifi cance was

reached at a p value below 0.05.

RESULTSHigh-fat feeding induces hepatic lipogenic gene expression in parallel to lipid accumulation

Mice fed the high-fat and high-fat/fi sh oil diet had higher caloric intakes (chow, 12.3±0.1; high-fat,

17.1±0.8, high-fat/fi sh oil, 16.8±1.0 kcal/day, p<0.05 high-fat vs. chow). At the end of the dietary

period, their body weight were increased compared to that of chow-fed animals (chow, 28.2±0.4;

high-fat, 30.5±0.8; high-fat/fi sh oil, 33.4±0.8 g, p<0.05 high-fat vs. chow and high-fat/fi sh oil vs. high-

fat). This was due to an increased fat mass (chow, 2.1±0.1; high-fat, 6.3±0.5; high-fat/fi sh oil, 7.6±0.4

% of total body weight, p<0.05 high-fat vs. chow). Expression of genes encoding enzymes involved

in hepatic fatty acid (Acc, Fas, Scd1, Elovl6) and TG (Dgat, Gpat) synthesis was induced in mice fed the

high-fat diet compared to animals receiving chow. Furthermore, Srebp-1c and Pgc-1β expression

was higher in these mice as compared to controls (Figure 1). The increase in lipogenic gene expres-

sion was associated with increases in hepatic TG (chow, 9.2±1.6; high-fat, 16.4±2.4 μmol/g, p<0.05)

CE (chow, 0.9±0.2; high-fat, 2.5±0.2 μmol/g, p<0.05) and free cholesterol (chow, 6.0±0.2; high-fat,

7.8±0.5 μmol/g, p<0.05) contents in mice fed the high-fat diet. Partial replacement of the saturated

fat within the high-fat diet by n-3 PUFA strongly suppressed hepatic lipogenic gene expression (Fig-

ure 1) and normalized hepatic TG and CE contents (Figure 2A and 2B) to values observed in chow-

fed mice. Hepatic phospholipid contents (Figure 2C and 2D) and liver weights (chow, 0.96±0.07;

high-fat, 1.08±0.04; high-fat/fi sh oil, 1.08±0.03 g) were similar in all mice. Hepatic fatty acid profi les

are shown in Table 1. The total amount of fatty acids was increased in mice fed the high-fat diet

compared to chow-fed controls due to accumulation of TG and CEs. In general, the contribution of

MUFAs was increased by high-fat feeding and desaturation indices were consequently increased.

Fish oil normalized hepatic MUFA content and the desaturation indices.

Srebp-1c

Pgc-1β

Acc1 Acc2 FasElovl6 Scd1

Dgat1Dgat2 Gpat

0

1

2

3

4


Rela

tive

mRN

A ex

pres

sion

##

##

#

# #

#

##

*

**

*

*

* *

Figure 1. Hepatic lipogenic gene expressions in C57Bl/6 mice fed chow, high-fat and high-fat/fi sh oil diets for 6 weeks.

White bars represent chow diet; black bars represent high-fat diet and grey bars represent high-fat/fi sh oil diet. Values represent means

± SEM for n=6/7; * p<0.05 high-fat vs. chow; # p<0.05 high-fat/fi sh oil vs. high-fat (Conover test).

9 9Chapter 6

0

10

20

30


Hepa

ticTG

(μm

ol/g

)

*

#

0

5

10

15


Hepa

tic fr

ee c

hole

ster

ol (μ

mol

/g)

*

0

1

2

3

4

5 chowhigh-fathigh-fat/fish oil

Hepa

tic C

E (μ

mol

/g)

*

#

0

20

40

60


Hepa

tic p

hosp

holip

ids

(μm

ol/g

)

A B

C D

Figure 2. Hepatic lipid content in C57Bl/6 mice fed chow, high-fat and high-fat/fi sh oil diets for 6 weeks.

A, Hepatic triglyceride content. B, Hepatic cholesterol ester content. C, Hepatic free cholesterol content. D, Hepatic phospholipid con-

tent.

White bars represent chow diet; black bars represent high-fat diet and grey bars represent high-fat/fi sh oil diet. Values represent means

± SEM for n=6/7; * p<0.05 high-fat vs. chow; # p<0.05 high-fat/fi sh oil vs. high-fat (Conover test).

Figure 3. Hepatic fatty acid synthesis in C57Bl/6 mice fed chow, high-fat and high-fat/fi sh oil diets for 6 weeks.

A, Fractional palmitate synthesis from de novo lipogenesis. B, Absolute palmitate synthesis from de novo lipogenesis. C, Fractional stearate

(C18:0) and oleate (C18:1) synthesis from elongation of de novo synthesized (C16:0DNL) and pre-existing (C16:0PE) palmitate. D, Absolute

stearate (C18:0) and oleate (C18:1) synthesis from elongation of de novo synthesized (C16:0DNL) and pre-existing (C16:0PE) palmitate.

White bars represent chow diet, black bars represent high-fat diet and grey bars represent high-fat/fi sh oil diet. Plain bars represent syn-

thesis from elongation of de novo synthesized palmitate and dashed bars represent synthesis from elongation of pre-existing palmitate.

Values represent means ± SEM for n=5-7; * p<0.05 high-fat vs. chow; # p<0.05 high-fat/fi sh oil vs. high-fat (Conover test).

0

10

20


Frac

tiona

l C16

:0 s

ynth

esis

(%)

#

0

10

20

chow C16:0 DNLhigh-fat C16:0 DNLhigh-fat/fish oil C16:0 DNL

Frac

tiona

l C18

syn

thes

is (%

)

C18:0 C18:1

chow C16:0 PE high-fat C16:0 PEhigh-fat/fish oil C16:0 PE

#

##

#

0

100


Abso

lute

C16

:0 s

ynth

esis

(μg/

g/h)

#

0

100

200chow C16:0 DNLhigh-fat C16:0 DNLhigh-fat/fish oil C16:0 DNL

Abso

lute

C18

syn

thes

is (μ

g/g/

h)

C18:0 C18:1

chow C16:0 PEhigh-fat C16:0 PEhigh-fat/fish oil C16:0 PE

#

#

##

*

*

* #

A B

C D

1 0 0 Dietary fatty acids alter lipogenic fl uxes

Table 2. Acetyl-CoA precursor pool enrichments in C57Bl/6 mice fed chow, high-fat or high-fat/fi sh oil diets for 6 weeks.


C16:0 8.8±0.5 13.0±0.3* 8.4±0.3#

C16:1 10.0±0.8 13.2±0.6* 10.2±1.2#

C18:0 5.7±0.2 8.1±0.5* 4.4±0.3#

C18:1 4.5±0.3 6.7±0.7* 7.0±1.0

Values represent means ± SEM for n=5-7 and expressed in percentages; * p<0.05 high-fat vs. chow; # p<0.05 high-fat/fi sh oil vs. high-fat

(Conover test).

High-fat feeding increases hepatic fatty acid synthesis from chain elongation

To assess whether the accumulation of hepatic lipids in response to high-fat feeding resulted from

an increased de novo fatty acid synthesis and/or chain elongation of de novo synthesized versus

existing palmitate, we infused [1-13C]-acetate and applied MIDA to the measured label distribution

patterns of palmitate, stearate and oleate, assuming that label incorporation was due to de novo

lipogenesis only. The resulting estimations of the acetyl-CoA precursor pool enrichments are shown

in Table 2. Compared to chow-fed animals, acetyl-CoA pool enrichments were increased in mice

fed the high-fat diet for all fatty acids analyzed. Precursor pool enrichments were similar in mice

fed chow and fi sh oil. In general, there was a clear discrepancy in acetyl-CoA pool enrichments for

palmitate (C16:0) and palmitoate (C16:1) on one hand and stearate (C18:0) and oleate (C18:1) on the

other hand. The precursor pool enrichment calculated for C16-fatty acids was higher than that cal-

culated to C18-fatty acids. This indicates that singly labeled fatty acids were high compared to triple

labeled fatty acids. We interpreted this diff erence as a refl ection of the diff erent synthetic pathways

of these fatty acids. C16-fatty acids are mainly synthesized by de novo lipogenesis, while C18-fatty

acids result from elongation of palmitate. Palmitate can either be synthesized de novo, or originate

from pre-existing sources. Accordingly, additional single labeled stearate is synthesized from elonga-

tion of pre-existing palmitate with a labeled acetyl-CoA moiety. This results in an excess contribu-

tion of single labeled molecules in C18-fatty acids over what could be anticipated based on the

contribution of triple labeled molecules originating from elongation of de novo synthesized palmi-

tate. The precursor pool enrichment, calculated by MIDA from C18-fatty acids will consequently be

underestimated compared to that calculated from C16-fatty acids. Therefore, we modifi ed the MIDA

algorithms [36–38] to account for excess single labeled C18-fatty acids, as described in Experimental

Procedures.

Compared to chow-fed animals, average fractional and absolute C16:0 synthesis were increased

by high-fat feeding, although the diff erence did not reach statistical signifi cance (Figure 3A and 3B).

High-fat feeding did not alter fractional and absolute C18:0 synthesis, and the contributions of de

novo synthesis and chain elongation were similar compared to chow-fed mice (Figure 3C and 3D).

Although high-fat feeding did not aff ect fractional C18:1 synthesis, the absolute synthesis by elon-

gation of both de novo synthesized (+300% vs. chow, p<0.05) and pre-existing palmitate (+213%

vs. chow, p<0.05) was increased as a result of the larger pool size. However, the contribution of

elongation of pre-existing palmitate to the increase in C18:1 synthesis was much more pronounced

1 0 1Chapter 6

compared to elongation of de novo synthesized palmitate (89 vs. 11%) in high-fat fed mice. When the

saturated fat was partially replaced by fi sh oil, fractional and absolute C16:0 synthesis were both sup-

pressed in high fat-fed mice (-66% and -70% vs. high-fat, p<0.05, Figure 3A and 3B). Furthermore, fi sh

oil inhibited C18:0 synthesis from elongation of both de novo synthesized and pre-existing palmitate

(total absolute C18:0 synthesis: -48% vs. high-fat, p<0.05). Strikingly, C18:1 synthesis from elongation

of de novo synthesized and pre-existing palmitate was almost completely abolished in fi sh oil-fed

mice (absolute C18:1 synthesis, -91% and -89% vs. high-fat, p<0.05, Figure 3C and 3D).

High-fat feeding increases cholesterol synthesis

High-fat feeding resulted in higher mRNA levels for enzymes involved in cholesterol biosynthesis

(i.e. Hmgcs1 and Hmgr) while expression of Acat1 and Acat2 was not aff ected (Figure 4A). Expression

of Srebp-2, which encodes a transcriptional regulator of cholesterol synthesis, was also induced. We

therefore determined fractional cholesterol synthesis in vivo following [1-13C]-acetate infusion and

MIDA. Again, high-fat feeding increased acetyl-CoA precursor pool enrichment (chow, 5.3±0.4; high-

fat, 10.2±0.8%, p<0.05). Moreover, high-fat feeding increased fractional cholesterol synthesis com-

pared to chow-fed controls (chow, 5.8±0.4; high-fat, 8.1±0.6%, p<0.05, Figure 4B). Fish oil normalized

mRNA expression of cholesterogenic genes, acetyl-CoA precursor pool enrichment (high-fat/fi sh oil,

6.0±0.4%, p<0.05 vs. high-fat) and fractional cholesterol synthesis (high-fat/fi sh oil, 5.6±0.3%, p<0.05

vs. high-fat, Figure 4B).

High-fat feeding does not aff ect hepatic VLDL-TG secretion

To assess whether high-fat feeding modulated hepatic lipid secretion, we determined VLDL-TG pro-

duction rates. Plasma TG levels prior to Poloxamer-407 injection were somewhat higher in mice fed

the high-fat diet compared to animals fed chow and lower in fi sh oil-fed mice (chow, 0.4±0.0; high-

fat, 0.5±0.0; high-fat/fi sh oil, 0.3±0.0 mM, p<0.05 chow vs. high-fat, high-fat vs. high-fat/fi sh oil, Figure

5A). High-fat feeding resulted in a slight statistically non-signifi cant reduction in hepatic VLDL-TG

production compared to chow-fed animals (chow, 168±8; high-fat, 154±6 μmol/kg/hour). In addi-

tion, the relative TG content of VLDL was decreased at the expense of phospholipids in mice fed the

high-fat diet (Table 3). Calculated size of nascent VLDL was reduced by high-fat feeding and West-

ern blot analysis revealed that high-fat feeding increased the relative amount of apoB48-containing

VLDL particles compared to mice fed chow (Table 3). Fish oil suppressed hepatic VLDL-TG produc-

tion (110±5 μmol/kg/hour, p<0.05 vs. high-fat, Figure 5A and 5B) and partially normalized the com-

position and size of the nascent VLDL particles to values observed in chow-fed animals (Table 3). In

addition, fi sh oil partially restored the balance between apoB48 and apoB100-containing particles.


Table 3. VLDL composition and calculated size in C57Bl/6 mice fed chow, high-fat or high-fat/fi sh oil diets for 6 weeks.


TG (mol%) 76.8±0.3 68.0±1.0* 73.1±0.4#

Phospholipids (mol%) 17.6±0.4 26.7±1.0* 19.0±0.4#

Cholesterol (mol%) 5.6±0.3 5.3±0.2 7.6±0.2#

Particle diameter (nm) 71.6±1.3 48.8±1.7* 64.2±1.2#

Particle volume (105 nm3) 1.9±0.1 0.6±0.1* 1.4±0.1#

ApoB48 (%) 80±5 92±1 86±2#

ApoB100 (%) 20±5 8±1 14±2#

Values represent means ± SEM for n=7-8 for the particle composition and size data and means ± SEM for n=4 for the apoB abundance;

* p<0.05 high-fat vs. chow; # p<0.05 high-fat/fi sh oil vs. high-fat (Conover test).

0 60 120 180 2400

5

10

15


Time (minutes)

Plas

ma

TG (m

M)

*#

#

##

0

100

200

300


VLDL

-TG

prod

uctio

n ra

te (μ

mol

/kg/

h)

#*

A B

Figure 5. Hepatic VLDL secretion in C57Bl/6 mice fed chow, high-fat and high-fat/fi sh oil diets for 6 weeks.

A, Plasma TG concentrations and B, VLDL-TG production rate.

White bullets and bars represent chow diet, black bullets and bars represent high-fat diet and grey bullets and bars represent high-fat/

fi sh oil diet. Values represent means ± SEM for n=7-8; * p<0.05 high-fat vs. chow; # p<0.05 high-fat/fi sh oil vs. high-fat (Conover test).

Srebp-2

Hmgsc1 HmgrAcat

1Acat

20

1

2

3


Rela

tive

mRN

A ex

pres

sion

# # ## #

*

*

*

0

5

10


Frac

tiona

l cho

lest

erol

syn

thes

is (%

)

#

*

A B

Figure 4. Cholesterol metabolism in C57Bl/6 mice fed chow, high-fat and high-fat/fi sh oil diets for 6 weeks.

A, Hepatic cholesterogenic gene expression. Cholesterogenic gene expression levels were calculated relative to the expression of 18S

and normalized for expression levels of control mice on chow. B, Fractional synthesis rates.

White bars represent chow diet, black bars represent high-fat diet and grey bars represent high-fat/fi sh oil diet. Values represent means

± SEM for n=5-7; * p<0.05 high-fat vs. chow; # p<0.05 high-fat/fi sh oil vs. high-fat (Conover test).

1 0 3Chapter 6

DISCUSSIONThe major fi nding of our study is that the counterintuitive induction of hepatic lipogenic genes

upon high-fat feeding is paralleled by adaptive remodelling of hepatic fatty acids rather than to

increased de novo lipogenesis. High-fat feeding also promoted cholesterol synthesis but did not

stimulate VLDL-TG secretion. Consequently, TG and CE accumulated in the livers of high fat-fed mice.

Partial replacement of the saturated fat for fi sh oil normalized hepatic lipid content by suppressing

both de novo lipogenesis and chain elongation as well as cholesterol synthesis.

We applied a novel approach based on in vivo 13C-acetate incorporation followed by MIDA to de-

termine the contribution of de novo lipogenesis and chain elongation to the synthesis of three major

hepatic fatty acids. The most commonly used method to determine hepatic lipogenesis in vivo in

experimental animals is by quantifi cation of the incorporation of 3H from 3H2O into total hepatic fatty

acids. However, this method only provides a rough estimate of fractional hepatic fatty acid synthesis

since 3H2O-derived label is incorporated into multiple positions in fatty acids by diff erent metabolic

pathways. Our approach provides more detailed information about the origin of the newly synthe-

sized fatty acids. We modifi ed the model introduced by Hellerstein and Neese [36–38] to determine

the contributions of chain elongation of de novo synthesized and pre-existing palmitate to stearate

and oleate synthesis. In the original model, palmitate synthesis is considered as a 8-step polymeriza-

tion of acetate units. Infusion of labeled acetate in vivo will result in its incorporation into fatty acids.

The frequency of label incorporation depends on the enrichment of the acetate pool, i.e. the precur-

sor pool enrichment. The newly synthesized fatty acids will either be labeled or unlabeled. This pool

of newly synthesized fatty acids is subsequently diluted in the existing pool of unlabeled fatty acids.

Thus, due to the synthesis of both labeled and unlabeled fatty acids, one cannot calculate fractional

fatty acid synthesis rates from the dilution of the labeled fatty acids only. Firstly, the enrichment of

the acetate pool is calculated from the M3/M

1 ratio, which is insensitive towards dilution. Secondly,

this precursor pool enrichment is used to calculate the theoretical (i.e. undiluted) frequency of triple

labeled fatty acids. This step in the MIDA actually also accounts for the synthesis of unlabeled fatty

acids. The ratio of the theoretical frequency over the measured amplitude of a particular fatty acid

mass isotopomer subsequently generates the dilution of the newly synthesized fatty acid. We ap-

plied this procedure to calculate fractional palmitate synthesis, assuming that this is solely refl ects

de novo lipogenesis. Next, we assumed that acetyl-CoA used for elongation of de novo synthesized

and pre-existing palmitate originates from the same pool, i.e., that the precursor pool enrichment

calculated for palmitate equals that for stearate and oleate. Finally, the mass isotopomer distribution

patterns of stearate and oleate were used to calculate the contributions of elongation of de novo

synthesized and pre-existing palmitate. Some studies have casted doubt on the homogeneity of

the hepatic acetyl-CoA pool [194,195], which could explain the diff erence in pool enrichments in

palmitate versus cholesterol observed in the current study. However, Hellerstein et al., have shown

that the enrichment of the precursor pool for fatty acid synthesis is very similar to that of acetate

residues in acetylated drugs [196]. Furthermore, inhomogeneous labeling of the hepatic acetyl-CoA

pool has been observed at very high degrees of labeling, i.e., around 70% [195]. We and others

[38,197–199] have avoided this issue by using protocols that result in a moderate precursor pool

labeling of ~15%.


The high fat diet-induced increase in lipogenic gene expression observed in this study, confi rms

earlier reports [130,185]. We now show that the increase in lipogenic genes does not result in a sig-

nifi cant induction of de novo lipogenesis, hence this pathway appears to be of minor physiological

importance in the development of hepatic lipid accumulation under conditions of high-fat feeding.

De novo synthesis of palmitic acid was not signifi cantly increased. Furthermore, the contribution

of elongation of de novo synthesized palmitate to absolute stearate and oleate synthesis was only

minor and represented 20 and 9% of the total synthesis, respectively, in chow-fed animals. A relative

decrease in the contribution of de novo lipogenesis to hepatic TG has recently also been reported

in rats fed a high-fat diet [200]. Moreover, de novo lipogenesis is not induced upon a short-term

dietary fat challenge in human subjects [201,202]. Palmitate synthesis from de novo lipogenesis may

however have been overestimated in the current study because we were not able to quantify the

contribution of chain elongation to the synthesis of this fatty acid. However, because of the relatively

low dietary myristic acid content, we consider this contribution to be of minor importance. Chain

elongation of pre-existing palmitate represented 89% of the increase in C18:1 synthesis upon high-

fat feeding. Another interesting fi nding in our study is the observation that partial eucaloric replace-

ment of saturated fat within the high-fat diet by fi sh oil completely abrogated the lipogenic eff ect.

Partial fi sh oil replacement was apparently suffi cient to normalize lipogenic gene expression profi les

and hepatic steatosis, even under conditions of inhibited VLDL-TG secretion. The suppressive eff ect

of n-3 PUFA on lipogenic gene expression in liver has been reported in earlier studies [189], however,

the physiological eff ect of fi sh oil on de novo lipogenesis and chain elongation in vivo has not been

investigated before. Our work shows that fi sh oil not only counteracts the increase in hepatic chain

elongation, but also suppresses fatty acid synthesis via the de novo lipogenic pathway.

Interestingly, lipid partitioning to TG storage has recently been suggested to protect the liver from

lipotoxicity [100]. Obesity and insulin resistance result in an increased fl ux of fatty acids from adipose

tissue towards the liver [180]. If hepatic fatty acid oxidation is not suffi cient to meet its infl ux, fatty ac-

ids may be elongated [203] and/or re-esterifi ed to prevent their toxic accumulation [2]. On the other

hand, high fat diet-induced steatosis is prevented if hepatic fatty acid infl ux is blocked by knockdown

of the fatty acid transporter FATP5 [204]. The increased fatty acid elongation and subsequent TG

synthesis upon high-fat feeding may therefore refl ect a physiological buff ering process. In addition,

increased hepatic fatty acid content induces ACAT activity [205,206], thereby promoting fatty acid

esterifi cation to cholesterol, as refl ected by higher hepatic CE contents in high fat-fed mice. As a con-

sequence, cellular free cholesterol content drops, which, in turn, provokes a compensatory SREBP-

2-mediated induction of cholesterol synthesis [207,208] that is refl ected by an increased expression

of cholesterogenic genes and increased cholesterol synthesis. Similar adaptive mechanisms leading

to CE accumulation exist when mice that are unable to exert feedback-inhibition by cellular choles-

terol (i.e., Lxrα -/- mice) are challenged with dietary cholesterol [68]. Thus, in response to an increased

substrate supply, the liver exerts several adaptive physiological responses to prevent cytotoxic ac-

cumulation of lipid species. In the current study, the increase in oleate synthesis most likely refl ects

the liver’s attempt to safely store saturated fatty acids as relatively harmless TGs. Palmitate was fi rst

elongated into stearate, which in turn was desaturated to oleate via SCD1 action. Stearate itself is

only minimally present in TGs [68]. Hepatic SCD1 action actually plays a key role in the partitioning of

excess lipid and enables adequate storage [100]. It has to be noted that the infl ux of dietary saturat-

1 0 5Chapter 6

ed fatty acids may have been limited, since we performed the experiments after a 4-hour fast. Under

these conditions, fatty acid infl ux from adipose tissue probably dominates. Obesity and insulin resist-

ance [180] may therefore in fact indirectly result in the increase in hepatic oleate synthesis via chain

elongation of circulating palmitate in high fat-fed mice. It should also be noted that the lipogenic

fl uxes maximally accounted for 15% of the hepatic fatty acid pool. Re-esterifi cation of circulating

fatty acids therefore represents the major pathway contributing to hepatic TG disposal. The induc-

tion of protective systems upon high-fat feeding is absent in case of partial isocaloric replacement of

the saturated fat by fi sh oil. In addition to a profound suppression of lipogenic gene expression [189],

such dietary modulation apparently alters the cellular fate of fatty acids and their infl ux into the liver.

In this respect, it should be noted that n-3 PUFA not only promote fatty acid oxidation [143], but also

increase peripheral lipid clearance presumably by enhancing LPL-activity [209–212]. As a result, fatty

acids are sequestered in extrahepatic tissues, predominantly in adipose stores. Mice fed the fi sh-oil

containing high-fat diet indeed deposited more fat in their adipose tissue.

Despite increased substrate availability and elevated hepatic Pgc-1β expression [185], hepatic

VLDL-TG production rate was slightly but non-signifi cantly reduced upon high-fat feeding. Hepatic

VLDL production was thus insuffi cient to accommodate the increase in hepatic fatty acid synthesis

during high fat feeding, indicating that there was progressive steatosis in high fat-fed mice. Moreo-

ver, VLDL particle size was reduced in high fat-fed mice and the relative abundance of apoB48-asso-

ciated VLDL particles was increased. The molecular mechanism underlying the decreased VLDL-TG

secretion rate is not yet clear. We can, however, conclude from this study that the rate of hepatic fatty

acid synthesis per se does not determine hepatic VLDL-TG secretion as proposed earlier [197,198,213]

and, thus, that VLDL-TG secretion is not determined by TG availability. Other factors such as mobiliza-

tion of the cytosolic TG pool and apoB availability and –fusion are therefore likely to be important

controlling factors in hepatic VLDL secretion [214–216]. Indeed, the suppression of VLDL secretion

by fi sh oil has been reported to be due to increased apoB degradation [217].

In summary, data reported in this study provide insight in a physiological mechanism that pro-

tects the liver from lipotoxicity under conditions of dietary fatty acid oversupply. Using a novel MIDA

approach we were able to show that high-fat feeding predominantly promotes fatty acid elongation

of pre-existing palmitate in vivo. Although high-fat feeding resulted in an induction of hepatic ex-

pression of de novo lipogenic genes, we did not observe a signifi cant increase in the fl ux through this

pathway. Furthermore, cholesterol synthesis is increased, presumably to compensate for increased

cholesterol esterifi cation. These physiological adaptations result in hepatic lipid accumulation and

do not occur if fatty acid infl ux into the liver is arrested by partial replacement of saturated fat by

fi sh oil. It should however be noted that the ‘harmless’ storage of excess fatty acids represents the

primary event or the ‘fi rst hit’ in the pathophysiology of NASH [181]. Consequently, such an adap-

tive physiological response may eventually predispose to development of liver disease, because it

renders the liver more susceptible to ‘second hits’ [181].

ACKNOWLEDGEMENTSThe authors thank Klaas Bijsterveld, Ingrid A. Martini, Claude P. van der Ley and Hermi Kingma for

excellent technical assistance.

7General Discussion

1 0 9Chapter 7

The fl ow of metabolic intermediates through biochemical pathways (‘metabolic fl uxes’) are to a cer-

tain extent controlled by ‘nutrient sensors’. These sensors enable adequate adaptation to changes in

nutrient availability and therefore contribute to the maintenance of energy homeostasis. Transcrip-

tion factors comprise a subgroup of nutrient sensors. Chronic energy oversupply and nutritional dys-

balance evoke persistent modifi cations in physiological processes. These adaptive responses may in

the long term predispose to the development of metabolic diseases.

Ligand-activated nuclear receptors are master regulators of whole-body metabolism: they control

the expression of genes encoding enzymes that constitute biochemical pathways. As such, they

are considered as putative drug targets to correct metabolic disturbances such as dyslipidemia and

insulin resistance. Insight into the body’s adaptive physiological responses to changes in nutrient

availability, and the role of transcriptional regulators herein, is needed to defi ne optimal strategies

for disease prevention and treatment.

Research described in this dissertation addresses the metabolic consequences of changes in nu-

trient availability. Stable isotope methodology was applied to quantify the actual metabolic fl uxes in

vivo and outcome was related to biochemical and gene expression analysis in relevant organs and

tissues.

NOVEL INSIGHTS INTO THE ACTION OF NUTRIENT-SENSING TRANSCRIPTION FACTORS Studies described in this dissertation provide new insights into the action of a specifi c set of tran-

scription factors.

Oxysterol activation of LXRs promotes cellular cholesterol disposal, by inducing the expression

of genes encoding transporters and enzymes that mediate cholesterol effl ux, cholesterol excretion

as well as cholesterol conversion into bile acids [218]. Furthermore, it has been shown that both

glucose and G6P are able to bind to and activate hepatic LXR at physiological concentrations in vitro

[53]. This issue has, however, been heavily debated [65–67]. In Chapter 2 we have tested the physi-

ological relevance of the postulated hepatic glucose sensing function of LXR in mice. We found that

the induction of lipogenic genes in liver and the increase of VLDL-TG concentrations upon carbo-

hydrate refeeding observed in wild-type mice were markedly blunted in Lxrα-/- mice. However, we

did not observe any eff ect of either carbohydrate refeeding or Lxrα disruption on the expression of

the LXR target genes Abca1 and Abcg5/8. The disruption of Lxrα did furthermore not aff ect hepatic

and peripheral insulin sensitivity, thereby confi rming previous studies [60]. The blunted lipogenic

response in carbohydrate-refed Lxrα-/- mice was therefore most likely related to an impaired SREBP-

1c action. We also noticed that the hepatic response to fasting was hampered in Lxrα-/- mice. Hepatic

G6P turnover was reduced and glycogen depletion was delayed. Fasting-induced steatosis was also

markedly less pronounced in these animals. In contrast to the impaired lipogenic induction upon

refeeding, the reduction in fasting-induced steatosis most likely results from the absence of Lxrα per

se, since fasted Srebp-1c -/- mice accumulated similar amounts of hepatic TG as compared to their

wild-type littermates [21]. Because hepatic LXRα was found to be insensitive to dietary glucose,

we hypothesize that the impaired hepatic response to fasting in Lxrα-/- mice may rather be related

to other metabolic changes associated with fasting, such as reduced energy availability and/or in-

1 1 0 Metabolic consequences of altered transcription factor action

creased NEFA infl ux and catabolism. Interestingly, LXRα has been implicated in the regulation of

adipose tissue lipolysis [219]. However, the absence of LXRα may also increase RXR availability for

other nuclear receptors, such as PPARα [220]. Increased PPARα action consequently promotes NEFA

catabolism [13,14,102]. Thus, instead of its anticipated involvement in the regulation of hepatic me-

tabolism in response to glucose and insulin [53,57] we have identifi ed LXRα as a mediator of the

adaptive response to fasting in the liver. This has recently been confi rmed by others (Sokolovic et al.,

personal communication).

FXRs are activated by bile acids. Upon activation, FXR suppresses bile acid synthesis while bile

acid disposal is promoted [221]. Hepatic FXR expression and transcriptional activity have also been

reported to be induced by glucose [88] and the hepatic response to short-term fasting is impaired

in Fxr -/- mice [44]. This has been suggested to result from an inadequate induction of gluconeogen-

esis. In Chapter 3, we report another feature of altered physiological responsiveness to changes in

glucose availability in Fxr -/- mice. The appearance of glucose entering the blood compartment dur-

ing the initial phase of (intestinal) glucose uptake was shown to be delayed in these animals. Using

a combination of orally and intravenously administered isotopically labeled glucose, we showed

that this delay was caused by an increased glucose fl ux through G6P in the enterocytes. Although

speculative, this may serve to restore the enterocyte’s depleted G6P stores observed in fasted Fxr -/-

mice (i.e., prior to the glucose load). Taken together, these data suggest that besides the incapability

to exert feedback regulation in response to increasing bile acid concentrations, FXR inactivation is

characterized by impaired sensing of a reduced glucose availability by liver and intestine. This pre-

sumably results from inadequate regulation of de novo synthesis and partitioning of G6P [44].

Activation of PPARα by fatty acids and their derivatives [12] results in an induction of genes en-

coding enzymes involved in their transport and catabolism [102–104]. PPARα has emerged as an

important mediator of the hepatic response to fasting by ensuring energy supply through fatty

acid oxidation when glucose availability is low [13,14]. In addition, PPARα protects against damage

from fatty acid oxidation products by promoting anti-oxidant action and mitochondrial uncoupling

[108,113,114]. In Chapter 4, we describe a novel metabolic consequence of pharmacological PPARα

activation. We observed a strong induction of genes encoding enzymes involved in hepatic fatty

acid synthesis and elongation in mice treated with the PPARα agonist fenofi brate. Using a novel

stable isotope approach, we quantifi ed the actual lipogenic fl ux and chain elongation in TG-derived

fatty acids. Both de novo lipogenesis and fatty acid elongation were massively induced upon PPARα

activation. Evaluation of hepatic carbohydrate fl uxes and gene expression levels indicated that

acetyl-CoA supply from glucose was reduced. Acetyl-CoA from fatty acid oxidation must therefore

have been the major lipogenic substrate. Using specifi c knockout mice treated with the pharma-

cological PPARα agonist, we found that the induction of lipogenic genes depended on SREBP-1c

but not on ChREBP. Srebp-1c expression itself was not induced upon pharmacological PPARα activa-

tion. The lipogenic induction must therefore have resulted from an increased SREBP-1c activity, as

has been proposed before [119]. Interestingly, PPARα agonists fail to induce SREBP-1c activity and

lipogenic gene expression in hepatoma cells [131] as well as in Pparα-/- mice [119]. These observa-

tions strongly suggest that an enhanced hepatic fatty acid infl ux (in response to PPARα/FGF-21-

mediated fatty acid mobilization) modifi es hepatic intracellular lipid status, which in turn promotes

SREBP-1c action. Although PPARα is an important regulator of the adaptive response to fasting, the

1 1 1Chapter 7

physiological consequences of pharmacological PPARα activation cannot be directly extrapolated

to the fasting situation, because it is generally accepted that SREBP-1c action is limited under these

conditions [222,223]. Furthermore, PPARα agonists induce both mitochondrial and peroxisomal fatty

acid oxidation, while during fasting, mitochondrial fatty acid oxidation predominates. Acetyl-CoA

from both sources has been reported to serve as a substrate for fatty acid synthesis [134,135]. The

relevance of the parallel existence of fatty acid oxidation and synthesis/elongation systems under

fasting conditions remains to be established. Altogether, our data support the co-existence of he-

patic β-oxidation and lipogenesis, thereby challenging the classical view that fat oxidation and syn-

thesis are two opposing biochemical processes that occur under diff erent metabolic conditions. We

propose the PPARα/SREBP-1c-mediated induction of hepatic fatty acid synthesis and elongation as

a novel physiological mechanism by which the liver is protected against fatty acids and their oxida-

tion products.

ADAPTIVE PHYSIOLOGAL RESPONSES TO A CHRONIC OVERLOAD OF DIETARY FAT Glucose and fatty acids exert direct substrate competition at the cellular level [11]. This normally

coincides with the availability of glucose during the postprandial and postabsorptive phases. An

increase in circulating NEFA concentrations acutely inhibits glucose utilization in vivo [224–226]. This

indicates that removal of circulating fatty acids and their subsequent oxidation occurs at the ex-

pense of glucose disposal.

We evaluated the eff ects of an increased dietary fat supply of on whole-body substrate utilization

(Chapter 5). Under normal conditions, carbohydrate oxidation provides the major energy supply in

the postprandial phase (~85%) while the contribution of fat oxidation increases during the postab-

sorptive state. In mice fed a regular low-fat chow diet, this is illustrated by a switch from carbohy-

drate oxidation (indicated by high RER values) during the dark phase (in which the animals are active

and consume most of their food) to fat oxidation (indicated by low RER values) during the light (or

inactive) phase. We observed that this physiological substrate switching was abolished when mice

were challenged with a hypercaloric high-fat diet (that still contained a considerable amount of

glucose). Instead, these animals exhibited a persistent reliance on fat oxidation. This phenotype was

even more pronounced when mice were fed a diet in which part of the saturated fat was isocalori-

cally replaced by fi sh oil. Fish oil is a source of n-3 PUFA. Compared to other types of fatty acids, n-3

PUFA exert a relatively high ability to bind PPARα and –δ [12], which may explain the additional reli-

ance on fat oxidation in fi sh oil-fed mice.

Consequences for glucose fl uxes

The reduced glucose-to-fat oxidation ratio in both high-fat and high-fat/fi sh oil-fed mice was par-

alleled by elevated plasma insulin concentrations, suggestive for insulin resistance. We therefore

evaluated the eff ects of these diets on whole-body glucose disposal and production in Chapter 5. Using hyperinsulinemic euglycemic clamps, we found that insulin-stimulated glucose disposal

was impaired in high fat-fed mice. Insulin’s ability to suppress endogenous glucose production was

also reduced in these animals. The additional decrease in the glucose-to-fat oxidation upon fi sh oil

replacement was associated with a further deterioration of insulin-stimulated glucose disposal.


Consequences for lipid fl uxes

Besides a sustained reliance on fat oxidation, we observed a counterintuitive induction of genes

encoding enzymes involved in fatty acid synthesis and elongation in the livers of high fat-fed mice

(Chapter 6), as had been reported by others [130,185]. The physiological relevance of these changes

had however not been established. Fish oil, on the other hand, is known to suppress the expression

of lipogenic genes by interfering with the action of the lipogenic transcription factors SREBP-1c, LXR

and ChREBP [20,29,30,33,227]. Fish oil replacement indeed reduced the expression levels to values

comparable to or below the chow-fed animals. We quantifi ed the actual lipogenic fl ux and deter-

mined the contributions of the de novo lipogenesis and chain elongation of pre-existing palmitate.

High-fat feeding resulted in a more than twofold increase in oleate synthesis. De novo synthesized

palmitate, however, only minimally contributed to this increase. Instead, elongation of pre-existing

palmitate mainly accounted for the induction of lipogenesis upon high-fat feeding. High-fat feed-

ing furthermore induced cholesterol synthesis, presumably to ensure cholesterol supply for fatty

acid esterifi cation. These adaptations resulted in the accumulation both of TGs and CEs, which was

progressive since there was no compensatory increase in hepatic VLDL-TG secretion. Fish oil replace-

ment of the saturated fat suppressed de novo lipogenic fl ux, almost completely abolished fatty acid

elongation and normalized cholesterol synthesis. As such, it prevented the high fat diet-induced

accumulation of TGs and CEs. It furthermore resulted in an inhibition of VLDL-TG secretion, which

contributed to the lowering of TG concentrations in the plasma. Although the hepatic lipogenic

gene expression profi les correlated well to the amount of hepatic TG and the total lipogenic fl uxes,

hepatic lipid accumulation in high-fat fed mice rather resulted from an increased fatty acid elonga-

tion than from an induction of de novo lipogenesis.

ENERGY OVERSUPPLY, NUTRITIONAL DYSBALANCE AND THEIR PATHOPHYSIOLOGICAL CONSEQUENCESObesity and lipid overfl ow

The body does not reduce its energy intake in response to fat oversupply; intestinal fatty acid ab-

sorption is actually increased in response to high-fat feeding in mice [228]. This response may serve

to limit exposure of enterocytes to NEFAs. Obesity results from a chronic imbalance between en-

ergy intake and energy expenditure, and is associated with an enhanced release of NEFA into the

circulation. The infl ux of fatty acids to organs is consequently increased. Such ‘lipid overfl ow’ from

adipose to non-adipose tissues is generally considered as a crucial event in the development of the

metabolic diseases.

Recent studies point to the storage capacity of the adipose tissue, and not the absolute amount

of fat per se, as an important determinant in the development of metabolic complications [229–231].

Lipid overfl ow may therefore not to be directly related to fat mass, but rather refl ects an inability to

store fat in adipocytes. This may explain why in some cases, the severity of complications in extreme

obese individuals is less than in those with a mild degree of obesity. Expansion of the fat storage

capacity in adipose tissue has been shown to improve insulin sensitivity [231,232], despite a con-

comitant increase in adiposity. On the other hand, an impaired ability to store fat in the adipose

tissue results in global metabolic failure [230,233,234].

1 1 3Chapter 7

Metabolic infl exibility and insulin resistance

Metabolic fl exibility is defi ned as the ability of a system to adjust fuel utilization to fuel availability.

The switch in fuel utilization will depend on the type and amount of nutrients available at the cel-

lular level [235]. Increased circulating NEFA levels will increase the amount of fatty acid available for

oxidation and subsequently impair glucose oxidation.

Physiological switching between carbohydrate and fat during active (fed) and inactive (fasting)

phases is abolished upon high-fat feeding in mice (Chapter 5). Such an adaptive shift in substrate

utilization is already observed upon acute exposure to high-fat diets (van den Berg et al., personal

communication), and thus occurs independently of obesity associated with long-term high-fat

feeding. It illustrates the dominance of fat oxidation over glucose oxidation, which may again refl ect

an attempt of the body to limit lipotoxic damage. This response, however, induces metabolic infl ex-

ibility to glucose. Because the high-fats also contain carbohydrates, blood glucose concentrations

will rise. This is illustrated by the hyperglycemia observed in high fat-fed mice. The elevated glucose

concentrations subsequently trigger insulin release, resulting in hyperinsulinemia. In addition, the

body’s responsiveness to insulin (i.e., insulin sensitivity) decreases, indicated by the impaired ability

of chronic hyperinsulinemia to restore glucose disposal rates in high fat-fed mice. This is presumably

related to diet-induced changes in membrane fatty acid composition aff ecting IR/IRS-PI3K action

[236–238]. IRS-dependent insulin signalling is furthermore reduced by an increase in intracellular

lipid species such as long-chain fatty acid-CoA, diacylglycerol and ceramide [239]. We indeed ob-

served an impairment in the insulin-mediated increase in IRS-associated PI3K activity in adipose

tissue and liver upon high-fat feeding (Chapter 5). The translocation of GLUT4 to the plasma mem-

brane is consequently impaired, and glucose uptake is reduced. Interestingly, a single high-glucose

low-fat meal has been reported to restore insulin-stimulated glucose disposal in skeletal muscles of

rats maintained on a high-fat diet during three weeks [236,240].

Human studies support the hypothesis of metabolic infl exibility to glucose as an initial event in

the pathogenesis of insulin resistance and type 2 diabetes. The ability to switch from fat oxidation to

carbohydrate oxidation after a meal is impaired in the prediabetic (i.e., glucose intolerant) state [172]

and metabolic infl exibility is predominantly related to defective glucose disposal in type 2 diabetic

subjects [174]. Furthermore, it is partially reversed by weight loss [172,174]. Besides a reduction of ca-

loric intake, restoration of nutrient balance may therefore be an eff ective approach to improve glu-

cose tolerance and insulin sensitivity. This implies that fat intake should be restricted to ~30% of total

energy, while glucose should not be provided as simple sugars, but as complex carbohydrates.

Hepatic steatosis

Dietary fat oversupply and obesity increase the fl ux of NEFA towards the liver. This promotes hepatic

lipid accumulation, as re-esterifi cation of circulating NEFA comprises a major contribution to the he-

patic TG pool [241,242]. Work described in Chapter 6 provides new insights into the adaptive physi-

ological responses to an increased hepatic NEFA infl ux. We have shown that in high fat-fed mice,

the liver elongates pre-existing palmitate, which is subsequently desaturated to facilitate its storage

as TG, while VLDL-TG output remains unaltered. Although oleate synthesized via this pathway only

comprises a minor part of the total pool, fatty acid elongation may signifi cantly contribute to steato-

sis progression in the long term. The relevance of this fi nding is illustrated by the observation that a

high-fat diet increases liver fat content in human subjects without aff ecting body fat [182]. High-fat


feeding furthermore promotes cholesterol synthesis, which results in the accumulation of hepatic

CEs.

Reduction of dietary fat intake and weight loss limit the hepatic NEFA infl ux and consequently ar-

rest hepatic lipid accumulation [182,243]. Consumption of fi sh oil promotes fat oxidation, thereby

reducing the amount of fatty acids available for hepatic re-esterifi cation. Specifi c types of fatty acids

furthermore aff ect LPL-mediated chylomicron clearance [211], thereby reducing fatty acid spill over

from chylomicron-derived TGs to the blood compartment [244]. The intake of n-3 PUFA may thus

sequester dietary fatty acids in adipose stores [209,210], thereby reducing lipid storage in the liver.

Finally, expansion of the adipocyte’s fat storage capacity upon TZD treatment improves liver func-

tion in NAFLD patients [245], presumably by preventing lipid overfl ow. Long-term continuation of

TZD therapy is required to maintain improvements in morbidity [246]. This is however accompanied

by undesired weight gain.

HEPATIC LIPID SYNTHESIS AS AN ADAPTIVE PHYSIOLOGICAL RESPONSE TO AN INCREASED FATTY ACID INFLUX. A ROLE FOR TRANSCRIPTION FACTORS?Hepatic storage of NEFA and acetyl-CoA from β-oxidation as relatively harmless TGs may represent

a physiological mechanism to prevent lipotoxic damage (Chapter 4 and 6; [230,247]). SCD1 action

appears to play a crucial role in this protective action, because it is required for the partitioning of

excess NEFA into MUFA, which in turn can be safely stored as TGs [100]. Hepatic Scd1 expression is

induced upon high-fat feeding in mice (Chapter 6 and [100,130]) and in human fatty liver [248].

Inhibition of SCD1 action increases susceptibility to saturated fat-induced apoptosis. Hepatocellular

Impaired glucose toleranceImpaired glucose toleranceImpaired glucose toleranceInsulin resistanceInsulin resistanceInsulin resistance

HyperinsulinemiaHyperinsulinemiaHyperinsulinemia

Reduced glucose oxidationReduced glucose oxidationReduced glucose oxidationPerturbed insulin signalingPerturbed insulin signalingPerturbed insulin signaling

Increased fat availabilityIncreased faff t availabilityIncreased fat availability

Sustained fat oxidationSustained faff t oxidationSustained fat oxidation

HyperglycemiaHyperglycemiaHyperglycemia

Figure 1. Overview of the adaptive physiological events in the development of lipid-induced insulin resistance.

1 1 5Chapter 7

apoptosis, liver injury, and fi brosis are markedly increased in Scd1-/- mice challenged with a methio-

nine-choline-defi cient, while steatosis is decreased in these animals [100]. This supposed protective

SCD1 action is not restricted to the liver [249].

The induction of hepatic SCD1 action in response to oversupply of dietary fat may be secondary

to increased transcription factor action. PPARs are likely candidates to mediate these adaptations,

because they are activated by fatty acids. Scd1 has actually been reported as a direct target gene

of PPARs [120,250]. Furthermore, Pparγ2 is ectopically induced in liver in response to overfeeding

[251,252]. PPARγ2 action has been reported to prevent lipotoxicity by facilitating the deposition as

fatty acids as TG: ablation of Pparγ2 in ob/ob mice results in an increased hepatic ceramide content,

in parallel to a reduction in TGs [230]. The studies described in Chapter 3 and 6 point to a regulatory

role for SREBP-1c, as has been suggested by others [185]. Evidence for an inadequate response to

high-fat feeding in the absence of SREBP-1c is however not available. Experiments involving Srebp-

1c -/- mice will provide more insight into the potential role of this transcription factor in the develop-

ment of high-fat diet induced hepatic steatosis.

Interestingly, the lipogenic transcriptional regulators PPARγ2, LXR and SREBP-1c [46–48] have all

been reported to be over-expressed in human fatty liver. Hepatic upregulation of transcription fac-

tor expression may therefore refl ect an attempt to protect the liver against fatty acid oversupply.

It should however be noted that rather harmless TG accumulation may eventually predispose to

development of liver disease, because it renders the liver more susceptible to ‘second hits’ [181].

TG

TG

TG

TG

NEFA

diet, adipose tissue

SRE

SREBP

elongation / desaturation

PPRE

CA

RXRE

CACACA

RXRPPAR

TG

Figure 2. Proposed mechanism by which an increased fl ux of fatty acids towards the liver promotes hepatic lipogenesis and TG storage

via the action of nutrient-sensing transcription factors.


CONCLUSIONS AND IMPLICATIONSThe metabolic syndrome comprises by a number of disturbances in energy and nutrient metabo-

lism including obesity, insulin resistance and dyslipidemia. Progression of these disturbances will

ultimately lead to type 2 diabetes and cardiovascular disease. The underlying pathophysiological

mechanisms are multifactorial. Prevention and/or treatment of this multimorbidity requires a global

approach, including the simultaneous modulation of multiple metabolic processes.

Transcription factors adjust the expression of metabolic enzymes in response to changes in nu-

trient availability. A change in the availability of metabolic enzymes may consequently aff ect the

fl ux through a biochemical pathway. The possibility to modulate metabolic fl uxes via the action of

ligand-activated nuclear receptors has sparked the interest to design drugs that act on these regula-

tors. This requires insight into the physiological consequences of transcription factor action.

When considering a nuclear receptor as a potential drug target, one should be aware that meta-

bolic remodelling may provoke undesired side-eff ects. An alteration of a specifi c biochemical reac-

tion will aff ect the fl ux through the entire pathway. In addition, the eff ect of a specifi c enzyme on the

global fl ux may furthermore vary under diff erent metabolic conditions [52]. Changes in the expres-

sion and/or abundance of metabolic enzymes may thus not always truly refl ect the actual metabolic

fl uxes, and the physiological consequences of transcription factor action may therefore be diff erent

than what is predicted from gene expression patterns. One should also be careful to draw general

conclusions on transcription factor action based on experimental evidence obtained under specifi c

conditions. In vivo evaluation of drug action is furthermore required to obtain a complete physi-

ological picture. This is of particular importance for multidrug approaches required to prevent and/

or treat the multimorbidities that comprise the metabolic syndrome.

These issues underline the need to apply fl uxomics in vivo, thereby establishing the physiological

relevance of a ‘static snaphot’ obtained from genomic, proteomic and metabolomic approaches

[35,253]. We have combined genomics, metabolics and fl uxomics to add to the current understand-

ing on the regulation of metabolic fl uxes by specifi c transcription factors. The results will contribute

to the development of new drugs to prevent and/or treat metabolic disturbances such as dyslipi-

demia and insulin resistance. However, in most cases, transcription factors are expressed in multi-

ple organs. Global targeting of these regulators may consequently induce unwanted physiological

responses. Dissection of the tissue-specifi c actions of transcription factors is of particular impor-

tance to identify undesirable side-eff ects of drug treatment. Therefore, possibilities for tissue-specifi c

targeting have to be explored and optimized. In addition, the application of metabolic pathway

analysis is of particular importance to test the usefulness of potential therapeutic strategies, and to

discover novel drug targets [254,255].

1 1 7Chapter 7


Supplemental Material

References

Summary

Samenvatting

Dankwoord

Biografi e/Biography

List of Publications

1 2 1Abbreviations


ABC ATP binding cassette LXR liver x receptor

ACAT acyl-CoA:cholesterol acyltransferase MIDA mass isotopomer distribution analysis

ACL ATP citrate lyase ME malic enzyme

ACC acetyl-CoA carboxylase MLX max-like protein x

apoB apolipoprotein B MUFA monounsaturated fatty acid

ATP adenosine triphosphate NADPH nicotinamide adenine dinucleotide phosphate

CACT carnitine-acylcarnitine translocase NAFLD non-alcoholic fatty liver disease

CD36 fatty acid transporter NEFA non-esterifi ed fatty acid

CE cholesterol ester OGTT oral glucose tolerance test

ChREBP carbohhydrate responsive element binding protein 6PDGH 6-phosphogluconate dehydrogenase

CPT carnitine palmitoyl transferase PDH pyruvate dehydrogenase

CRAT carnitine acyltransferase PDK pyruvate dehydrogenase kinase

DGAT diacylglycerol acyltransferase PEPCK phosphoenolpyruvate carboxykinase

EGP endogenous glucose production PGC-1 ppar gamma co-activator 1

ER endoplasmatic reticulum PI3K phosphoinositide-3 kinase

FAS fatty acid synthase PK pyruvate kinase

FBP1 fructose-1,6-biphosphatase 1 PPAR peroxisome proliferator activated receptor

FGF fi broblast growth factor PP2A protein phosphatase 2A

FXR farnesoid x receptor PPP pentose phosphate pathway

GC-MS gas chromatography-mass spectometry PUFA polyunsaturated fatty acid

GLUT glucose transporter qPCR quantitative PCR

G6P glucose-6-phosphate Ra rate of appearance

G6Pase glucose-6-phosphatase Rd rate of disappearance

G6Pdh glucose-6-phosphate dehydrogenase RE response element

G6Ph glucose-6-phosphate hydrolase RER respiratory exchange ratio

G6Pt glucose-6-phosphate translocase RXR retinoid x receptor

GK glucokinase SCAP srebp cleavage activating protein

GP glycogen phosphorylase SCD1 stearoyl-CoA desaturase 1

GS glycogen synthase SREBP sterol regulatory element binding protein

GPAT glycerol-3-phosphate acyltransferase TALDO transaldolase

β-HB β-hydroxybutyrate TCA tricarboxylic acid

HK hexokinase TG triglyceride

HMGCS 3-hydroxy-3-methylglutaryl-CoA synthase TKT transketolase

HMGR 3-hydroxy-3-methylglutaryl-CoA reductase TZD thiazolidinedione

IR insulin receptor UCP uncoupling protein

IRS insulin receptor substrate VLDL very low density lipoprotein

LPL lipoprotein lipase

1 2 3Supplemental Material

Supplemental Figure 1Schematic model of hepatic carbohydrate metabolism

triose phosphate

glucose glucose-6-phosphate glycogen

[U -13C][U - C]-glucose [1 -2H]-galactose

1

2

5

4

3

[2 -13C]-glycerol

blood sample glucose

urine sample paracetamol glucuronic acid

UDP-glucose

paracetamol

Major metabolic pathways and enzymatic reactions are depicted, sharing G6P as a central metabolite. The pathways included are:

(1) Gluconeogenic fl ux toward G6P, (2) Glycogen phosphorylase fl ux, (3) Glucose-6-phosphatase fl ux, (4) Glucokinase fl ux and (5)

Glycogen synthase fl ux. Mice received an infusion containing [U-13C]glucose, [2-13C]glycerol, [1-2H]galactose and paracetamol for six

hours. MIDA was applied on blood glucose and urinary paracetamol glucuronide samples.


Supplemental Table 1Primer and probe sequences used for qPCR

Gene Sense Antisense Probe Accession number

Acc1 CCA TCC AAA CAG AGG GAA CAT C CTA CAT GAG TCA TGC CAT AGT GGT T

ACG CTA AAC AGA ATG TCC TTT

GCC TCC AAC NM_133360.2

Acc2 CCC AGG AGG CTG CAT TGA AGA CAT GCT GGG CCT CAT AGT A

CAC AAG TGA TCC TGA ATC TCA

CGC GC NM_133904.1

Acs GGA GCT TCG CAG TGG CAT C CCC AGG CTC GAC TGT ATC TTG T

CAG AAA CAA CAG CCT GTG GGA

TAA ACT CAT CTT NM_007981

Aox GCC ACG GAA CTC ATC TTC GA CCA GGC CAC CAC TTA ATG GA

CCA CTG CCA CAT ATG ACC CCA

AGA CCC NM_015729

Elovl5 TGG CTG TTC TTC CAG ATT GGA CCC TTT CTT GTT GTA AGT CTG AAT GTA

CAT GAT TTC CCT GAT TGC TCT CTT

CAC AAA C NM_134255.2

Elovl6 ACA CGT AGC GAC TCC GAA GAT AGC GCA GAA AAC AGG AAA GAC T

TTT CCT GCA TCC ATT GGA TGG

CTT C NM_130450.2

Fads1 CCT TCG CGG ACA TTG TTT ACT C TAT GGA GGT CTG CTG CTG CTA T

CTC TGG TTG GAC GCT TAC CTT

CAC CA NM_013402.3

Fads2 CCC TGA TCG ACA TTG TGA GTT C GAC GGC AGC TTC ATT TAT GGA

CCA GCC ACA GCT CCC CAG

ACT TCT NM_019699.1

Fgf-21 CCG CAG TCC AGA AAG TCT CC TGA CAC CCA GGA TTT GAA TGA C

CCT GGC TTC AAG GCT TTG AGC

TCC A NM_020013.4

G6pdh GCA ACA GAT ACA AGA ATG TGA AGC T AGG CTT CCC TGA GTT CAT CAC T

CCT ATG AAC GCC TCA TCC TGG

ATG TCT T NM_008062

Gyk GGG TTG GTG TGT GGA GTC TTG GAT TTC GCT TTC TTC AGC ATT GA

ACC GCT CCA TTG TGA CAG

CTG ACA NM_008194.3

Hmgcs1 CGA TGG TGT AGA TGC TGG AAA G CAT CAG TTT CTG AAC CAC AGT CGA TCC GTG CAG AAG CCC ATC C NM_145942.2

Lcad TAC GGC ACA AAA GAA CAG ATC G CAG GCT CTG TCA TGG CTA TGG CAC TTG CCC GCC GTC ATC TGG NM_007381

Me1 AGG CAG CGT CTT CCA AAT ATG TCG ATA CTT GTT CAG GAG ACG AA

TGG CAA AAT CTT CAA ACT GAA

TAA GGC AAT TC NM_008615.1

6Pdgh GGA CAT CCG TAA GGC CCT CTA T ATT GAG GGT CCA GCC AAA CTC

CTT TAT GCT GCT CAG ACA GGC

AGC CAC NM_025801

Pgc-1β GAG ACA CAG ATG AAG ATC CAA GCT CTT GCC AAG AGA GTC GCT TTG T CCA GGT GCC TCA TGC TGG CCT NM_133249

Sglt1 GTT GGA GTC TAC GCA ACA GCA A GGG CTT CTG TGT CTA TTT CAA TTG T TCC TCC TCT CCT GCA TCC AGG TCG NM_019810.3

Taldo1 CAG AAG TTG ATG CAA GGC TTT C CCA GCT TCT TTG TAA AGC TCG A CTC GGG CCA CCA TGG CAT CC NM_011528

Tkt GCA TCC TGT CCC GAA ACA AG CAA TAG ACT CGG TAG CTG GCT TT CCC TGG CCC AGG GAG CCA GT NM_009388

1 2 6 Supplemental Material

Supplemental Table 2Parameters and equations used to calculate hepatic glucose metabolism

Parameter Equation

Primary isotopic parameters

1. d(glc) M6(glc)blood/M

6(glc)infusate

2. d(UDPglc) M1(pGlcUA)urine/M

1(gal)infusate

3. c(glc) M6(pGlcUA)urine/M

6(glc)blood

4. c(UDPglc) M1(glc)blood/M

1(pGlcUA)urine

5. f(glc) M2(glc)blood/M

2(FBP:MIDA)glc

6. f(UDPglc) M2(pGlcUA)urine/M

2(FBP:MIDA)pGlcUA

Whole body glucose metabolism

7. Ra(glc;whole body) Inf(glc;total)/d(glc)

8. Ra(UDPglc;whole body) Inf(gal;total)/d(UDPglc)

9. MCR(glc) Ra(glc;whole body)/glc conc

10. Ra(glc;endo) Ra(glc;whole body) - Inf(gal;total)

11. Ra(UDPglc;endo) Ra(UDPglc;whole body) - Inf(gal;total)

12. Rr(glc) {c(glc)/[1-c(glc)]}/Ra(glc;endo)

13. Rr(UDPglc) {c(UDPglc)/[1-c(UDPglc)]}/Ra(UDPglc;endo)

14. Total Ra(glc;endo) Ra(glc;endo) + Rr(glc)

15. Total Ra(UDPglc;endo) Ra(UDPglc;endo) + Rr(UDPglc)

16. UDPglc(glc) c(UDPglc) x [Ra(glc;endo) + Inf(glc;total)]

17. glc(UDPglc) c(glc) x [Ra(UDPglc;endo) + Inf(gal;total)]

18. GNG(glc) f(glc) x [Ra(glc;whole body) + Rr(glc)]

19. GNG(UDPglc) f(UDPglc) x [Ra(UDPglc;whole body) + Rr(UDPglc)]

20. GNG(glc;indirect) [f(UDPglc) x UDPglc(glc)] + [f(glc] x Rr(glc)]

21. GNG(UDPglc;indirect) [f(glc) x glc(UDPglc)] + [f(UDPglc) x Rr(UDPglc)]

22. GNG(glc;direct) GNG(glc) - GNG(glc;indirect)

23. GNG(UDPglc;direct) GNG(UDPglc) - GNG(UDPglc;indirect)

24. GLY(glc) Ra(glc;endo) - GNG(glc;direct) - [f(UDPglc) x UDPglc(glc)]

25. GLY(UDPglc) Ra(UDPglc;endo) - GNG(UDPglc;direct) - glc(UDPglc)

Individual fl uxes comprising hepatic G6P metabolism

26. GNG(G6P) GNG(glc;direct) + GNG(UDPglc;direct)

27. GK glc(UDPglc) + Rr(glc)

28. G6Pase GNG(glc) + GLY(glc)

29. GS GNG(UDPglc) + GLY(UDPglc)

30. GP GLY(UDPglc) + GLY(glc) + {[ 1- c(glc)] x Rr(UDPglc)}


d(glc) fractional contribution of infused glu-

cose to blood glucose

glc conc blood glucose concentration in mM

M6(glc)infusate mole percent enrichments (MPE) of

[U-13C]glucose in the infusate

Ra(glc;endo) rate of endogenous blood glucose

appearance, not corrected for recy-

cling of tracer

M6(glc)blood MPE of [U-13C]glucose in blood Ra(UDPglc;endo) rate of endogenous UDP glucose ap-

pearance, not corrected for recycling

of tracer

d(UDPglc) fractional contribution of infused ga-

lactose to UDPglucose

Rr(glc) rate of recycling of glucose tracer

M1(gal)infusate MPE of [1-2H]galactose in the infusate Rr(UDPglc) rate of recycling of UDPglc tracer

M1(pGlcUA)urine MPE of [1-2H]-UDPglucose measured

in urinary Par-GlcUA

totalRa(glc;endo) total endogenous glucose produc-

tion, including recycling of tracer

c(glc) fractional contribution of blood glu-

cose to UDP-glucose formation

totalRa(UDPglc) total endogenous UDPglucose pro-

duction, including recycling of tracer

M6(pGlcUA)urine MPE of [U-13C]- UDPglucose measured


UDPglc(glc) rate of UDPglucose conversion into

blood glucose

c(UDPglc) fractional contribution of UDPglucose

to blood glucose formation

glc(UDPglc) rate of blood glucose conversion into

UDPglucose

M1(glc)blood MPE of [1-2H]- glucose in blood GNG(glc) rate of gluconeogenesis into blood

glucose

M6(pGlcUA)urine MPE of [U-13C]-UDPglucose measured


GNG(UDPglc) rate of gluconeogenesis into UDP-

glucose

f(glc) fractional contribution of newly syn-

thesized glucose to blood glucose

GNG(glc;indirect) rate of gluconeogenesis into blood

glucose indirectly via glycogen

M2(glc)blood MPE of [13C

2]-glucose in blood GNG(UDPglc;indirect) rate of gluconeogenesis into UDPglu-

cose indirectly via blood glucose

M2(FBP:MIDA)glc theoretical MPE of [13C

2]-Fructose 1,6

biphosphate, calculated by MIDA

using 13C-enrichment data of glucose

GNG(glc;direct) rate of gluconeogenesis directly into

blood glucose

f(UDPglc) fractional contribution of newly syn-

thesized glucose to UDPglc pool

GNG(UDPglc;direct) rate of gluconeogenesis directly into

UDPglucose

M2(pGlcUA)urine MPE of [13C

2]-UDPglc , sampled as uri-

nary Par-GlcUA

GLY(glc) rate of glycogenolysis contributing to

blood glucose formation

M2(FBP:MIDA)pGlcUA theoretical MPE of [13C

2]-Fructose 1,6

biphosphate, calculated by MIDA

using 13C enrichment data of urinary

Par-GlcUA

GLY (UDPglc) rate of glycogenolysis contributing to

UDPglucose formation

Ra(glc;whole body) whole body rate of appearance of glu-

cose into the blood glucose pool

GNG(G6P) total fl ux of G6P de novo synthesis,

corrected for the exchange between

blood glucose and UDPglucose

pools

Inf(glc) rate of infusion of [U-13C]glucose in

μmol/kg/min

GK glucokinase fl ux

Ra(UDPglc;

whole body)

whole body rate of appearance of

UDPglc

G6Pase glucose-6-phosphatase fl ux

Inf(gal) rate of infusion of [1-2H]galactose in

μmol/kg/min

GS glycogen synthase fl ux

MCR(glc) metabolic clearance rate of blood

glucose

GP glycogen phosphorylase fl ux

1 2 8 Supplemental Material

Supplemental Table 3Formulas used to calculate the concentration vs. time curves and

the kinetic parameters in a fi rst order absorption process in an one-compartment model

Parameter Equation

Oral [U-13C]-glucose blood concentration Ct = M

6 x [glc]

[U-13C]-glucose in blood over time Ct = C(0)el x e-k(el) x t – C(0)ab x e-k(ab) x t

Lag time tlag

= (ln(C(0)ab) – ln(C(0)el))/(kab – kel)

[U-13C]-glucose blood concentration at tlag Clag

= C(0)el x e-k(el) x t(lag) = C(0)ab x e-k(ab) x t(lag)

Curve of absorption and elimination Ct = C

lag x (e-k(el) x (t-t(lag)) - e-k(ab) x (t-t(lag)))

Time were curve reaches its maximum tmax

= 1/(kab – kel) x ln(kab/kel)

[U-13C]-glucose blood concentration at tmax Cmax

= Clag

x (e-k(el) x t(max) – e-k(ab) x t(max))

Half life t½ = ln(2)/kel

Mean residence time MRT = 1/kab + 1/kel

Bioavailability F = 1 – ((C(0)ab x kel)/(kab x C(0)el))

Volume of distribution VD = F x D

L / C

lag

C(0)ab initial concentration by extrapolation of the absorption period

C(0)el concentration by extrapolation of the elimination period

Clag

concentration at lag time calculated from elimination or absorption curve

Cmax

concentration at tmax

DL

oral dose administrated

F fractional contribution to the sampled compartment

[glc] total blood glucose concentration

kab absorption rate constant

kel elimination rate constant

M6

mole percent enrichment of blood glucose

MRT mean residence time in sampled compartment

t(½)el half-life

tlag

time between administration and appearance in sampled compartment

tmax

time of maximal concentration

VD

appearant volume of distribution


Supplemental Table 4Fatty acid composition of the experimental diets


C14:0 0.5 12.2 16.1

C16:0 8.4 92.5 79.5

C16:1 0.7 11.5 18.0

C18:0 3.7 76.3 50.5

C18:1 13.7 133.2 101.0

C18:2 16.9 11.5 9.7

C18:3 1.9 2.9 15.2

C20-22 0.4 4.0 53.3

Values are given in g/kg.

1 3 0 References

ReferencesMithieux G, Gautier-Stein A, Rajas F, Zitoun C. 2006. Contribution of intestine and kidney to glucose fl uxes in dif-1.

ferent nutritional states in rat. Comp Biochem Physiol B Biochem Mol Biol 143: 195-200.

Malhi H, Gores GJ. 2008. Molecular mechanisms of lipotoxicity in nonalcoholic fatty liver disease. Semin Liver Dis 2.

28: 360-369.

Su X, Abumrad NA. 2009. Cellular fatty acid uptake: a pathway under construction. Trends Endocrinol Metab 20: 3.

72-77.

Ehehalt R, Fullekrug J, Pohl J, Ring A, Herrmann T, Stremmel W. 2006. Translocation of long chain fatty acids across 4.

the plasma membrane--lipid rafts and fatty acid transport proteins. Mol Cell Biochem 284: 135-140.

Furuhashi M, Hotamisligil GS. 2008. Fatty acid-binding proteins: role in metabolic diseases and potential as drug 5.

targets. Nat Rev Drug Discov 7: 489-503.

Kim JJ, Miura R. 2004. Acyl-CoA dehydrogenases and acyl-CoA oxidases. Structural basis for mechanistic similari-6.

ties and diff erences. Eur J Biochem 271: 483-493.

Jakobsson A, Westerberg R, Jacobsson A. 2006. Fatty acid elongases in mammals: their regulation and roles in 7.

metabolism. Prog Lipid Res 45: 237-249.

Pereira SL, Leonard AE, Mukerji P. 2003. Recent advances in the study of fatty acid desaturases from animals and 8.

lower eukaryotes. Prostaglandins Leukot Essent Fatty Acids 68: 97-106.

van Ginneken, V, Verhey E, Poelmann R, Ramakers R, van Dijk KW, Ham L, Voshol P, Havekes L, Van Eck M, van der 9.

Greef J. 2007. Metabolomics (liver and blood profi ling) in a mouse model in response to fasting: a study of hepatic

steatosis. Biochim Biophys Acta 1771: 1263-1270.

Jensen MD, Ekberg K, Landau BR. 2001. Lipid metabolism during fasting. Am J Physiol Endocrinol Metab 281: 10.

E789-E793.

Randle PJ, Garland PB, Hales CN, Newsholme EA. 1963. The glucose fatty-acid cycle. Its role in insulin sensitivity 11.

and the metabolic disturbances of diabetes mellitus. Lancet 1: 785-789.

Forman BM, Chen J, Evans RM. 1997. Hypolipidemic drugs, polyunsaturated fatty acids, and eicosanoids are li-12.

gands for peroxisome proliferator-activated receptors alpha and delta. Proc Natl Acad Sci U S A 94: 4312-4317.

Kersten S, Seydoux J, Peters JM, Gonzalez FJ, Desvergne B, Wahli W. 1999. Peroxisome proliferator-activated recep-13.

tor alpha mediates the adaptive response to fasting. J Clin Invest 103: 1489-1498.

Leone TC, Weinheimer CJ, Kelly DP. 1999. A critical role for the peroxisome proliferator-activated receptor alpha 14.

(PPARalpha) in the cellular fasting response: the PPARalpha-null mouse as a model of fatty acid oxidation disor-

ders. Proc Natl Acad Sci U S A 96: 7473-7478.

Tontonoz P, Spiegelman BM. 2008. Fat and beyond: the diverse biology of PPARgamma. Annu Rev Biochem 77: 15.

289-312.

Barish GD, Narkar VA, Evans RM. 2006. PPAR delta: a dagger in the heart of the metabolic syndrome. J Clin Invest 16.

116: 590-597.

Furnsinn C, Willson TM, Brunmair B. 2007. Peroxisome proliferator-activated receptor-delta, a regulator of oxidative 17.

capacity, fuel switching and cholesterol transport. Diabetologia 50: 8-17.

Hong C, Tontonoz P. 2008. Coordination of infl ammation and metabolism by PPAR and LXR nuclear receptors. Curr 18.

Opin Genet Dev 18: 461-467.

Repa JJ, Mangelsdorf DJ. 2000. The role of orphan nuclear receptors in the regulation of cholesterol homeostasis. 19.

Annu Rev Cell Dev Biol 16: 459-481.

Yoshikawa T, Shimano H, Yahagi N, Ide T, Amemiya-Kudo M, Matsuzaka T, Nakakuki M, Tomita S, Okazaki H, Tamura 20.

Y, Iizuka Y, Ohashi K, Takahashi A, Sone H, Osuga Ji J, Gotoda T, Ishibashi S, Yamada N. 2002. Polyunsaturated fatty

acids suppress sterol regulatory element-binding protein 1c promoter activity by inhibition of liver X receptor

(LXR) binding to LXR response elements. J Biol Chem 277: 1705-1711.

1 3 1References

Liang G, Yang J, Horton JD, Hammer RE, Goldstein JL, Brown MS. 2002. Diminished hepatic response to fasting/21.

refeeding and liver X receptor agonists in mice with selective defi ciency of sterol regulatory element-binding

protein-1c. J Biol Chem 277: 9520-9528.

Lefebvre P, Cariou B, Lien F, Kuipers F, Staels B. 2009. Role of bile acids and bile acid receptors in metabolic regula-22.

tion. Physiol Rev 89: 147-191.

Pai JT, Guryev O, Brown MS, Goldstein JL. 1998. Diff erential stimulation of cholesterol and unsaturated fatty acid 23.

biosynthesis in cells expressing individual nuclear sterol regulatory element-binding proteins. J Biol Chem 273:

26138-26148.

Yang T, Espenshade PJ, Wright ME, Yabe D, Gong Y, Aebersold R, Goldstein JL, Brown MS. 2002. Crucial step in cho-24.

lesterol homeostasis: sterols promote binding of SCAP to INSIG-1, a membrane protein that facilitates retention

of SREBPs in ER. Cell 110: 489-500.

Horton JD, Shimomura I, Brown MS, Hammer RE, Goldstein JL, Shimano H. 1998. Activation of cholesterol syn-25.

thesis in preference to fatty acid synthesis in liver and adipose tissue of transgenic mice overproducing sterol

regulatory element-binding protein-2. J Clin Invest 101: 2331-2339.

Repa JJ, Liang G, Ou J, Bashmakov Y, Lobaccaro JM, Shimomura I, Shan B, Brown MS, Goldstein JL, Mangelsdorf DJ. 26.

2000. Regulation of mouse sterol regulatory element-binding protein-1c gene (SREBP-1c) by oxysterol receptors,

LXRalpha and LXRbeta. Genes Dev 14: 2819-2830.

Chen G, Liang G, Ou J, Goldstein JL, Brown MS. 2004. Central role for liver X receptor in insulin-mediated activation 27.

of Srebp-1c transcription and stimulation of fatty acid synthesis in liver. Proc Natl Acad Sci U S A 101: 11245-

11250.

Shimomura I, Shimano H, Korn BS, Bashmakov Y, Horton JD. 1998. Nuclear sterol regulatory element-binding 28.

proteins activate genes responsible for the entire program of unsaturated fatty acid biosynthesis in transgenic

mouse liver. J Biol Chem 273: 35299-35306.

Xu J, Nakamura MT, Cho HP, Clarke SD. 1999. Sterol regulatory element binding protein-1 expression is suppres-29.

sed by dietary polyunsaturated fatty acids. A mechanism for the coordinate suppression of lipogenic genes by

polyunsaturated fats. J Biol Chem 274: 23577-23583.

Kim HJ, Takahashi M, Ezaki O. 1999. Fish oil feeding decreases mature sterol regulatory element-binding protein 30.

1 (SREBP-1) by down-regulation of SREBP-1c mRNA in mouse liver. A possible mechanism for down-regulation of

lipogenic enzyme mrnas. J Biol Chem 274: 25892-25898.

Postic C, Dentin R, Denechaud PD, Girard J. 2007. ChREBP, a transcriptional regulator of glucose and lipid metabo-31.

lism. Annu Rev Nutr 27: 179-192.

Kawaguchi T, Osatomi K, Yamashita H, Kabashima T, Uyeda K. 2002. Mechanism for fatty acid "sparing" eff ect on 32.

glucose-induced transcription: regulation of carbohydrate-responsive element-binding protein by AMP-activa-

ted protein kinase. J Biol Chem 277: 3829-3835.

Dentin R, Denechaud PD, Benhamed F, Girard J, Postic C. 2006. Hepatic gene regulation by glucose and polyunsa-33.

turated fatty acids: a role for ChREBP. J Nutr 136: 1145-1149.

Brunengraber H, Kelleher JK, Des RC. 1997. Applications of mass isotopomer analysis to nutrition research. Annu 34.

Rev Nutr 17: 559-596.

Hellerstein MK. 2003. 35. In vivo measurement of fl uxes through metabolic pathways: the missing link in functional

genomics and pharmaceutical research. Annu Rev Nutr 23: 379-402.

Hellerstein MK, Neese RA. 1992. Mass isotopomer distribution analysis: a technique for measuring biosynthesis 36.

and turnover of polymers. Am J Physiol 263: E988-1001.

Neese RA, Faix D, Kletke C, Wu K, Wang AC, Shackleton CH, Hellerstein MK. 1993. Measurement of endogenous 37.

synthesis of plasma cholesterol in rats and humans using MIDA. Am J Physiol 264: E136-E147.

Hellerstein MK, Kletke C, Kaempfer S, Wu K, Shackleton CH. 1991. Use of mass isotopomer distributions in secreted 38.

lipids to sample lipogenic acetyl-CoA pool in vivo in humans. Am J Physiol 261: E479-E486.

1 3 2 References

Lee WN, Byerley LO, Bergner EA, Edmond J. 1991. Mass isotopomer analysis: theoretical and practical considerati-39.

ons. Biol Mass Spectrom 20: 451-458.

Van Dijk TH, Boer TS, Havinga R, Stellaard F, Kuipers F, Reijngoud DJ. 2003. Quantifi cation of hepatic carbohydrate 40.

metabolism in conscious mice using serial blood and urine spots. Anal Biochem 322: 1-13.

Oosterveer MH, Van Dijk TH, Grefhorst A, Bloks VW, Havinga R, Kuipers F, Reijngoud DJ. 2008. Lxralpha defi ciency 41.

hampers the hepatic adaptive response to fasting in mice. J Biol Chem 283: 25437-25445.

Galgani J, Ravussin E. 2008. Energy metabolism, fuel selection and body weight regulation. Int J Obes (Lond) 32 42.

Suppl 7: S109-S119.

Sugden MC, Bulmer K, Gibbons GF, Knight BL, Holness MJ. 2002. Peroxisome-proliferator-activated receptor-alpha 43.

(PPARalpha) defi ciency leads to dysregulation of hepatic lipid and carbohydrate metabolism by fatty acids and

insulin. Biochem J 364: 361-368.

Cariou B, van HK, Duran-Sandoval D, van DT, Grefhorst A, Bouchaert E, Fruchart JC, Gonzalez FJ, Kuipers F, Staels B. 44.

2005. Transient impairment of the adaptive response to fasting in FXR-defi cient mice. FEBS Lett 579: 4076-4080.

Cariou B, van Harmelen K, Duran-Sandoval D, Van Dijk TH, Grefhorst A, Abdelkarim M, Caron S, Torpier G, Fruchart 45.

JC, Gonzalez FJ, Kuipers F, Staels B. 2006. The farnesoid X receptor modulates adiposity and peripheral insulin

sensitivity in mice. J Biol Chem 281: 11039-11049.

Westerbacka J, Kolak M, Kiviluoto T, Arkkila P, Siren J, Hamsten A, Fisher RM, Yki-Jarvinen H. 2007. Genes involved in 46.

fatty acid partitioning and binding, lipolysis, monocyte/macrophage recruitment, and infl ammation are overex-

pressed in the human fatty liver of insulin-resistant subjects. Diabetes 56: 2759-2765.

Higuchi N, Kato M, Shundo Y, Tajiri H, Tanaka M, Yamashita N, Kohjima M, Kotoh K, Nakamuta M, Takayanagi R, 47.

Enjoji M. 2008. Liver X receptor in cooperation with SREBP-1c is a major lipid synthesis regulator in nonalcoholic

fatty liver disease. Hepatol Res 38: 1122-1129.

Mitsuyoshi H, Yasui K, Harano Y, Endo M, Tsuji K, Minami M, Itoh Y, Okanoue T, Yoshikawa T. 2009. Analysis of he-48.

patic genes involved in the metabolism of fatty acids and iron in nonalcoholic fatty liver disease. Hepatol Res 39:

366-373.

Staels B, Maes M, Zambon A. 2008. Fibrates and future PPARalpha agonists in the treatment of cardiovascular 49.

disease. Nat Clin Pract Cardiovasc Med 5: 542-553.

Cho N, Momose Y. 2008. Peroxisome proliferator-activated receptor gamma agonists as insulin sensitizers: from 50.

the discovery to recent progress. Curr Top Med Chem 8: 1483-1507.

Grefhorst A, Elzinga BM, Voshol PJ, Plosch T, Kok T, Bloks VW, van der Sluijs FH, Havekes LM, Romijn JA, Verkade HJ, 51.

Kuipers F. 2002. Stimulation of lipogenesis by pharmacological activation of the liver X receptor leads to produc-

tion of large, triglyceride-rich very low density lipoprotein particles. J Biol Chem 277: 34182-34190.

Rossell S, van der Weijden CC, Lindenbergh A, van TA, Francke C, Bakker BM, Westerhoff HV. 2006. Unraveling the 52.

complexity of fl ux regulation: a new method demonstrated for nutrient starvation in Saccharomyces cerevisiae.

Proc Natl Acad Sci U S A 103: 2166-2171.

Mitro N, Mak PA, Vargas L, Godio C, Hampton E, Molteni V, Kreusch A, Saez E. 2007. The nuclear receptor LXR is a 53.

glucose sensor. Nature 445: 219-223.

Willy PJ, Umesono K, Ong ES, Evans RM, Heyman RA, Mangelsdorf DJ. 1995. LXR, a nuclear receptor that defi nes a 54.

distinct retinoid response pathway. Genes Dev 9: 1033-1045.

Teboul M, Enmark E, Li Q, Wikstrom AC, Pelto-Huikko M, Gustafsson JA. 1995. OR-1, a member of the nuclear recep-55.

tor superfamily that interacts with the 9-cis-retinoic acid receptor. Proc Natl Acad Sci U S A 92: 2096-2100.

Lehmann JM, Kliewer SA, Moore LB, Smith-Oliver TA, Oliver BB, Su JL, Sundseth SS, Winegar DA, Blanchard DE, 56.

Spencer TA, Willson TM. 1997. Activation of the nuclear receptor LXR by oxysterols defi nes a new hormone res-

ponse pathway. J Biol Chem 272: 3137-3140.

Cha JY, Repa JJ. 2007. The liver X receptor (LXR) and hepatic lipogenesis. The carbohydrate-response element-57.

binding protein is a target gene of LXR. J Biol Chem 282: 743-751.

1 3 3References

Schultz JR, Tu H, Luk A, Repa JJ, Medina JC, Li L, Schwendner S, Wang S, Thoolen M, Mangelsdorf DJ, Lustig KD, 58.

Shan B. 2000. Role of LXRs in control of lipogenesis. Genes Dev 14: 2831-2838.

Yoshikawa T, Shimano H, Amemiya-Kudo M, Yahagi N, Hasty AH, Matsuzaka T, Okazaki H, Tamura Y, Iizuka Y, Ohashi 59.

K, Osuga J, Harada K, Gotoda T, Kimura S, Ishibashi S, Yamada N. 2001. Identifi cation of liver X receptor-retinoid

X receptor as an activator of the sterol regulatory element-binding protein 1c gene promoter. Mol Cell Biol 21:

2991-3000.

Grefhorst A, Van Dijk TH, Hammer A, van der Sluijs FH, Havinga R, Havekes LM, Romijn JA, Groot PH, Reijngoud DJ, 60.

Kuipers F. 2005. Diff erential eff ects of pharmacological liver X receptor activation on hepatic and peripheral insulin

sensitivity in lean and ob/ob mice. Am J Physiol Endocrinol Metab 289: E829-E838.

Laffi tte BA, Chao LC, Li J, Walczak R, Hummasti S, Joseph SB, Castrillo A, Wilpitz DC, Mangelsdorf DJ, Collins JL, Saez 61.

E, Tontonoz P. 2003. Activation of liver X receptor improves glucose tolerance through coordinate regulation of

glucose metabolism in liver and adipose tissue. Proc Natl Acad Sci U S A 100: 5419-5424.

Cao G, Liang Y, Broderick CL, Oldham BA, Beyer TP, Schmidt RJ, Zhang Y, Stayrook KR, Suen C, Otto KA, Miller AR, Dai 62.

J, Foxworthy P, Gao H, Ryan TP, Jiang XC, Burris TP, Eacho PI, Etgen GJ. 2003. Antidiabetic action of a liver x receptor

agonist mediated by inhibition of hepatic gluconeogenesis. J Biol Chem 278: 1131-1136.

Grempler R, Gunther S, Steff ensen KR, Nilsson M, Barthel A, Schmoll D, Walther R. 2005. Evidence for an indirect 63.

transcriptional regulation of glucose-6-phosphatase gene expression by liver X receptors. Biochem Biophys Res

Commun 338: 981-986.

Commerford SR, Vargas L, Dorfman SE, Mitro N, Rocheford EC, Mak PA, Li X, Kennedy P, Mullarkey TL, Saez E. 2007. 64.

Dissection of the insulin-sensitizing eff ect of liver x receptor ligands. Mol Endocrinol 21: 3002-3012.

Lazar MA, Willson TM. 2007. Sweet dreams for LXR. Cell Metab 5: 159-161.65.

Denechaud PD, Dentin R, Girard J, Postic C. 2008. Role of ChREBP in hepatic steatosis and insulin resistance. FEBS 66.

Lett 582: 68-73.

Denechaud PD, Bossard P, Lobaccaro JM, Millatt L, Staels B, Girard J, Postic C. 2008. ChREBP, but not LXRs, is re-67.

quired for the induction of glucose-regulated genes in mouse liver. J Clin Invest 118: 956-964.

van der Veen JN, Havinga R, Bloks VW, Groen AK, Kuipers F. 2007. Cholesterol feeding strongly reduces hepatic 68.

VLDL-triglyceride production in mice lacking the liver X receptor alpha. J Lipid Res 48: 337-347.

Blight EG, Dyer WJ. 1959. A rapid method of total lipid extraction and purifi cation. Can J Biochem Physiol 37: 69.

911-917.

Hohorst HJ. 1970. D-Glucose-6-phosphat und D-fructose-6-phosphat. In: Bergmeyer HU (eds) Methoden der En-70.

zymatischen Analyse. Verlag Chemie, Weinheim, Germany, pp. 1200-1204.

Keppler D, Decker K. 1970. Glykogen. Bestimmung mit amyloglucosidase. In: Bergmeyer HU (eds) Methoden der 71.

Enzymatischen Analyse. Verlag Chemie, Weinheim, Germany, pp. 1089-1094.

Kuipers F, Havinga R, Bosschieter H, Toorop GP, Hindriks FR, Vonk RJ. 1985. Enterohepatic circulation in the rat. 72.

Gastroenterology 88: 403-411.

Bandsma RH, Grefhorst A, Van Dijk TH, van der Sluijs FH, Hammer A, Reijngoud DJ, Kuipers F. 2004. Enhanced 73.

glucose cycling and suppressed de novo synthesis of glucose-6-phosphate result in a net unchanged hepatic

glucose output in ob/ob mice. Diabetologia 47: 2022-2031.

Derks TG, Van Dijk TH, Grefhorst A, Rake JP, Smit GP, Kuipers F, Reijngoud DJ. 2008. Inhibition of mitochondrial 74.

fatty acid oxidation in vivo only slightly suppresses gluconeogenesis but enhances clearance of glucose in mice.

Hepatology 47: 1032-1042.

Youn JH, Bergman RN. 1990. Enhancement of hepatic glycogen by gluconeogenic precursors: substrate fl ux or 75.

metabolic control? Am J Physiol 258: E899-E906.

Bandsma RH, Van Dijk TH, Harmsel At A, Kok T, Reijngoud DJ, Staels B, Kuipers F. 2004. Hepatic 76. de novo synthesis

of glucose 6-phosphate is not aff ected in peroxisome proliferator-activated receptor alpha-defi cient mice but is

preferentially directed toward hepatic glycogen stores after a short term fast. J Biol Chem 279: 8930-8937.

1 3 4 References

Xu J, Xiao G, Trujillo C, Chang V, Blanco L, Joseph SB, Bassilian S, Saad MF, Tontonoz P, Lee WN, Kurland IJ. 2002. 77.

Peroxisome proliferator-activated receptor alpha (PPARalpha) infl uences substrate utilization for hepatic glucose

production. J Biol Chem 277: 50237-50244.

Burgess SC, Leone TC, Wende AR, Croce MA, Chen Z, Sherry AD, Malloy CR, Finck BN. 2006. Diminished hepatic 78.

gluconeogenesis via defects in tricarboxylic acid cycle fl ux in peroxisome proliferator-activated receptor gamma

coactivator-1alpha (PGC-1alpha)-defi cient mice. J Biol Chem 281: 19000-19008.

Jungermann K, Kietzmann T. 1996. Zonation of parenchymal and nonparenchymal metabolism in liver. Annu Rev 79.

Nutr 16: 179-203.

Greenberg CC, Jurczak MJ, Danos AM, Brady MJ. 2006. Glycogen branches out: new perspectives on the role of 80.

glycogen metabolism in the integration of metabolic pathways. Am J Physiol Endocrinol Metab 291: E1-E8.

Kuipers F, Claudel T, Sturm E, Staels B. 2004. The Farnesoid X Receptor (FXR) as modulator of bile acid metabolism. 81.

Rev Endocr Metab Disord 5: 319-326.

Lu TT, Makishima M, Repa JJ, Schoonjans K, Kerr TA, Auwerx J, Mangelsdorf DJ. 2000. Molecular basis for feedback 82.

regulation of bile acid synthesis by nuclear receptors. Mol Cell 6: 507-515.

Chiang JY. 2002. Bile acid regulation of gene expression: roles of nuclear hormone receptors. Endocr Rev 23: 443-83.

463.

Inagaki T, Choi M, Moschetta A, Peng L, Cummins CL, McDonald JG, Luo G, Jones SA, Goodwin B, Richardson JA, 84.

Gerard RD, Repa JJ, Mangelsdorf DJ, Kliewer SA. 2005. Fibroblast growth factor 15 functions as an enterohepatic

signal to regulate bile acid homeostasis. Cell Metab 2: 217-225.

Sinal CJ, Tohkin M, Miyata M, Ward JM, Lambert G, Gonzalez FJ. 2000. Targeted disruption of the nuclear receptor 85.

FXR/BAR impairs bile acid and lipid homeostasis. Cell 102: 731-744.

Claudel T, Inoue Y, Barbier O, Duran-Sandoval D, Kosykh V, Fruchart J, Fruchart JC, Gonzalez FJ, Staels B. 2003. Farne-86.

soid X receptor agonists suppress hepatic apolipoprotein CIII expression. Gastroenterology 125: 544-555.

Claudel T, Sturm E, Duez H, Torra IP, Sirvent A, Kosykh V, Fruchart JC, Dallongeville J, Hum DW, Kuipers F, Staels 87.

B. 2002. Bile acid-activated nuclear receptor FXR suppresses apolipoprotein A-I transcription via a negative FXR

response element. J Clin Invest 109: 961-971.

Duran-Sandoval D, Mautino G, Martin G, Percevault F, Barbier O, Fruchart JC, Kuipers F, Staels B. 2004. Glucose 88.

regulates the expression of the farnesoid X receptor in liver. Diabetes 53: 890-898.

Yamagata K, Daitoku H, Shimamoto Y, Matsuzaki H, Hirota K, Ishida J, Fukamizu A. 2004. Bile acids regulate gluco-89.

neogenic gene expression via small heterodimer partner-mediated repression of hepatocyte nuclear factor 4 and

Foxo1. J Biol Chem 279: 23158-23165.

Ma K, Saha PK, Chan L, Moore DD. 2006. Farnesoid X receptor is essential for normal glucose homeostasis. J Clin 90.

Invest 116: 1102-1109.

Stumpel F, Burcelin R, Jungermann K, Thorens B. 2001. Normal kinetics of intestinal glucose absorption in the 91.

absence of GLUT2: evidence for a transport pathway requiring glucose phosphorylation and transfer into the

endoplasmic reticulum. Proc Natl Acad Sci U S A 98: 11330-11335.

Uldry M, Thorens B. 2004. The SLC2 family of facilitated hexose and polyol transporters. Pfl ugers Arch 447: 480-92.

489.

Higashiyama H, Kinoshita M, Asano S. 2008. Immunolocalization of farnesoid X receptor (FXR) in mouse tissues 93.

using tissue microarray. Acta Histochem 110: 86-93.

Kok T, Hulzebos CV, Wolters H, Havinga R, Agellon LB, Stellaard F, Shan B, Schwarz M, Kuipers F. 2003. Enterohe-94.

patic circulation of bile salts in farnesoid X receptor-defi cient mice: effi cient intestinal bile salt absorption in the

absence of ileal bile acid-binding protein. J Biol Chem 278: 41930-41937.

Cobelli C, Mari A, Ferrannini E. 1987. Non-steady state: error analysis of Steele's model and developments for 95.

glucose kinetics. Am J Physiol 252: E679-E689.

Mari A. 1992. Estimation of the rate of appearance in the non-steady state with a two-compartment model. Am 96.

J Physiol 263: E400-E415.

1 3 5References

Proietto J, Rohner-Jeanrenaud F, Ionescu E, Terrettaz J, Sauter JF, Jeanrenaud B. 1987. Non-steady-state measure-97.

ment of glucose turnover in rats by using a one-compartment model. Am J Physiol 252: E77-E84.

Steele E R. 1959. Infl uences of glucose loading and of injected insulin on hepatic glucose output. Ann N Y Acad 98.

Sci 82: 420-430.

Tissot S, Normand S, Guilluy R, Pachiaudi C, Beylot M, Laville M, Cohen R, Mornex R, Riou JP. 1990. Use of a new gas 99.

chromatograph isotope ratio mass spectrometer to trace exogenous 13C labelled glucose at a very low level of

enrichment in man. Diabetologia 33: 449-456.

Li ZZ, Berk M, McIntyre TM, Feldstein AE. 2009. Hepatic lipid partitioning and liver damage in nonalcoholic fatty 100.

liver disease: role of stearoyl-CoA desaturase. J Biol Chem 284: 5637-5644.

Patsouris D, Reddy JK, Muller M, Kersten S. 2006. Peroxisome proliferator-activated receptor alpha mediates the 101.

eff ects of high-fat diet on hepatic gene expression. Endocrinology 147: 1508-1516.

Mandard S, Muller M, Kersten S. 2004. Peroxisome proliferator-activated receptor alpha target genes. Cell Mol Life 102.

Sci 61: 393-416.

Inagaki T, Dutchak P, Zhao G, Ding X, Gautron L, Parameswara V, Li Y, Goetz R, Mohammadi M, Esser V, Elmquist 103.

JK, Gerard RD, Burgess SC, Hammer RE, Mangelsdorf DJ, Kliewer SA. 2007. Endocrine regulation of the fasting

response by PPARalpha-mediated induction of fi broblast growth factor 21. Cell Metab 5: 415-425.

Badman MK, Pissios P, Kennedy AR, Koukos G, Flier JS, Maratos-Flier E. 2007. Hepatic fi broblast growth factor 21 104.

is regulated by PPARalpha and is a key mediator of hepatic lipid metabolism in ketotic states. Cell Metab 5: 426-

437.

Dalen KT, Ulven SM, Arntsen BM, Solaas K, Nebb HI. 2006. PPARalpha activators and fasting induce the expression 105.

of adipose diff erentiation-related protein in liver. J Lipid Res 47: 931-943.

Martin PG, Guillou H, Lasserre F, Dejean S, Lan A, Pascussi JM, Sancristobal M, Legrand P, Besse P, Pineau T. 2007. 106.

Novel aspects of PPARalpha-mediated regulation of lipid and xenobiotic metabolism revealed through a nutrige-

nomic study. Hepatology 45: 767-777.

Devchand PR, Keller H, Peters JM, Vazquez M, Gonzalez FJ, Wahli W. 1996. The PPARalpha-leukotriene B4 pathway 107.

to infl ammation control. Nature 384: 39-43.

Kelly LJ, Vicario PP, Thompson GM, Candelore MR, Doebber TW, Ventre J, Wu MS, Meurer R, Forrest MJ, Conner 108.

MW, Cascieri MA, Moller DE. 1998. Peroxisome proliferator-activated receptors gamma and alpha mediate in vivo

regulation of uncoupling protein (UCP-1, UCP-2, UCP-3) gene expression. Endocrinology 139: 4920-4927.

Schrauwen P, Hesselink MK. 2004. The role of uncoupling protein 3 in fatty acid metabolism: protection against 109.

lipotoxicity? Proc Nutr Soc 63: 287-292.

Andrews ZB, Horvath TL. 2009. Uncoupling protein-2 regulates lifespan in mice. Am J Physiol Endocrinol Metab 110.

296: E621-E627.

Schrader M, Fahimi HD. 2006. Peroxisomes and oxidative stress. Biochim Biophys Acta 1763: 1755-1766.111.

Gueraud F, Alary J, Costet P, Debrauwer L, Dolo L, Pineau T, Paris A. 1999. 112. In vivo involvement of cytochrome

P450 4A family in the oxidative metabolism of the lipid peroxidation product trans-4-hydroxy-2-nonenal, using

PPARalpha-defi cient mice. J Lipid Res 40: 152-159.

Toyama T, Nakamura H, Harano Y, Yamauchi N, Morita A, Kirishima T, Minami M, Itoh Y, Okanoue T. 2004. PPARalpha 113.

ligands activate antioxidant enzymes and suppress hepatic fi brosis in rats. Biochem Biophys Res Commun 324:

697-704.

Inoue I, Noji S, Awata T, Takahashi K, Nakajima T, Sonoda M, Komoda T, Katayama S. 1998. Bezafi brate has an 114.

antioxidant eff ect: peroxisome proliferator-activated receptor alpha is associated with Cu2+, Zn2+-superoxide

dismutase in the liver. Life Sci 63: 135-144.

Stienstra R, Mandard S, Patsouris D, Maass C, Kersten S, Muller M. 2007. Peroxisome proliferator-activated receptor 115.

alpha protects against obesity-induced hepatic infl ammation. Endocrinology 148: 2753-2763.

1 3 6 References

Shiri-Sverdlov R, Wouters K, van Gorp PJ, Gijbels MJ, Noel B, Buff at L, Staels B, Maeda N, van BM, Hofker MH. 2006. 116.

Early diet-induced non-alcoholic steatohepatitis in APOE2 knock-in mice and its prevention by fi brates. J Hepatol

44: 732-741.

Yeon JE, Choi KM, Baik SH, Kim KO, Lim HJ, Park KH, Kim JY, Park JJ, Kim JS, Bak YT, Byun KS, Lee CH. 2004. Reduced 117.

expression of peroxisome proliferator-activated receptor-alpha may have an important role in the development

of non-alcoholic fatty liver disease. J Gastroenterol Hepatol 19: 799-804.

Seo YS, Kim JH, Jo NY, Choi KM, Baik SH, Park JJ, Kim JS, Byun KS, Bak YT, Lee CH, Kim A, Yeon JE. 2008. PPAR ago-118.

nists treatment is eff ective in a nonalcoholic fatty liver disease animal model by modulating fatty-acid metabolic

enzymes. J Gastroenterol Hepatol 23: 102-109.

Knight BL, Hebbachi A, Hauton D, Brown AM, Wiggins D, Patel DD, Gibbons GF. 2005. A role for PPARalpha in the 119.

control of SREBP activity and lipid synthesis in the liver. Biochem J 389: 413-421.

Miller CW, Ntambi JM. 1996. Peroxisome proliferators induce mouse liver stearoyl-CoA desaturase 1 gene expres-120.

sion. Proc Natl Acad Sci U S A 93: 9443-9448.

Iizuka K, Bruick RK, Liang G, Horton JD, Uyeda K. 2004. Defi ciency of carbohydrate response element-binding 121.

protein (ChREBP) reduces lipogenesis as well as glycolysis. Proc Natl Acad Sci U S A 101: 7281-7286.

Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. 1951. Protein measurement with the Folin phenol reagent. J Biol 122.

Chem 193: 265-275.

Muskiet FA, van Doormaal JJ, Martini IA, Wolthers BG, van der SW. 1983. Capillary gas chromatographic profi ling of 123.

total long-chain fatty acids and cholesterol in biological materials. J Chromatogr 278: 231-244.

Oosterveer MH, Van Dijk TH, Tietge UJ, Boer T, Havinga R, Stellaard F, Groen AK, Kuipers F, Reijngoud DJ. 2009. High 124.

fat feeding induces hepatic fatty acid elongation in mice. PLoS One 4: e6066.

Kabashima T, Kawaguchi T, Wadzinski BE, Uyeda K. 2003. Xylulose 5-phosphate mediates glucose-induced lipo-125.

genesis by xylulose 5-phosphate-activated protein phosphatase in rat liver. Proc Natl Acad Sci U S A 100: 5107-

5112.

Castelein H, Gulick T, Declercq PE, Mannaerts GP, Moore DD, Baes MI. 1994. The peroxisome proliferator activated 126.

receptor regulates malic enzyme gene expression. J Biol Chem 269: 26754-26758.

Hebbachi AM, Knight BL, Wiggins D, Patel DD, Gibbons GF. 2008. Peroxisome proliferator-activated receptor alpha 127.

defi ciency abolishes the response of lipogenic gene expression to re-feeding: restoration of the normal response

by activation of liver X receptor alpha. J Biol Chem 283: 4866-4876.

Chakravarthy MV, Pan Z, Zhu Y, Tordjman K, Schneider JG, Coleman T, Turk J, Semenkovich CF. 2005. "New" hepatic 128.

fat activates PPARalpha to maintain glucose, lipid, and cholesterol homeostasis. Cell Metab 1: 309-322.

Ogata M, Tsujita M, Hossain MA, Akita N, Gonzalez FJ, Staels B, Suzuki S, Fukutomi T, Kimura G, Yokoyama S. 2009. 129.

On the mechanism for PPAR agonists to enhance ABCA1 gene expression. Atherosclerosis 205: 413-419.

Sampath H, Miyazaki M, Dobrzyn A, Ntambi JM. 2007. Stearoyl-CoA desaturase-1 mediates the pro-lipogenic ef-130.

fects of dietary saturated fat. J Biol Chem 282: 2483-2493.

Konig B, Koch A, Spielmann J, Hilgenfeld C, Hirche F, Stangl GI, Eder K. 2009. Activation of PPARalpha and PPAR-131.

gamma reduces triacylglycerol synthesis in rat hepatoma cells by reduction of nuclear SREBP-1. Eur J Pharmacol

605: 23-30.

Lundasen T, Hunt MC, Nilsson LM, Sanyal S, Angelin B, Alexson SE, Rudling M. 2007. PPARalpha is a key regulator 132.

of hepatic FGF21. Biochem Biophys Res Commun 360: 437-440.

Cheung C, Akiyama TE, Ward JM, Nicol CJ, Feigenbaum L, Vinson C, Gonzalez FJ. 2004. Diminished hepatocellular 133.

proliferation in mice humanized for the nuclear receptor peroxisome proliferator-activated receptor alpha. Cancer

Res 64: 3849-3854.

Wong DA, Bassilian S, Lim S, Paul Lee WN. 2004. Coordination of peroxisomal beta-oxidation and fatty acid elon-134.

gation in HepG2 cells. J Biol Chem 279: 41302-41309.

1 3 7References

Reszko AE, Kasumov T, David F, Jobbins KA, Thomas KR, Hoppel CL, Brunengraber H, Des RC. 2004. Peroxisomal 135.

fatty acid oxidation is a substantial source of the acetyl moiety of malonyl-CoA in rat heart. J Biol Chem 279:

19574-19579.

Sugden MC, Holness MJ. 2003. Recent advances in mechanisms regulating glucose oxidation at the level of the 136.

pyruvate dehydrogenase complex by PDKs. Am J Physiol Endocrinol Metab 284: E855-E862.

Patsouris D, Mandard S, Voshol PJ, Escher P, Tan NS, Havekes LM, Koenig W, Marz W, Tafuri S, Wahli W, Muller M, 137.

Kersten S. 2004. PPARalpha governs glycerol metabolism. J Clin Invest 114: 94-103.

Xu J, Chang V, Joseph SB, Trujillo C, Bassilian S, Saad MF, Lee WN, Kurland IJ. 2004. Peroxisomal proliferator-acti-138.

vated receptor alpha defi ciency diminishes insulin-responsiveness of gluconeogenic/glycolytic/pentose gene

expression and substrate cycle fl ux. Endocrinology 145: 1087-1095.

Sriram G, Rahib L, He JS, Campos AE, Parr LS, Liao JC, Dipple KM. 2008. Global metabolic eff ects of glycerol kinase 139.

overexpression in rat hepatoma cells. Mol Genet Metab 93: 145-159.

Spolarics Z. 1998. Endotoxemia, pentose cycle, and the oxidant/antioxidant balance in the hepatic sinusoid. J 140.

Leukoc Biol 63: 534-541.

Riccardi G, Giacco R, Rivellese AA. 2004. Dietary fat, insulin sensitivity and the metabolic syndrome. Clin Nutr 23: 141.

447-456.

Carpentier YA, Portois L, Malaisse WJ. 2006. n-3 fatty acids and the metabolic syndrome. Am J Clin Nutr 83: 1499S-142.

1504S.

Davidson MH. 2006. Mechanisms for the hypotriglyceridemic eff ect of marine omega-3 fatty acids. Am J Cardiol 143.

98: 27i-33i.

Sirtori CR, Galli C. 2002. N-3 fatty acids and diabetes. Biomed Pharmacother 56: 397-406.144.

Rivellese AA, Lilli S. 2003. Quality of dietary fatty acids, insulin sensitivity and type 2 diabetes. Biomed Pharma-145.

cother 57: 84-87.

Mostad IL, Bjerve KS, Bjorgaas MR, Lydersen S, Grill V. 2006. Eff ects of n-3 fatty acids in subjects with type 2 diabe-146.

tes: reduction of insulin sensitivity and time-dependent alteration from carbohydrate to fat oxidation. Am J Clin

Nutr 84: 540-550.

Giacco R, Cuomo V, Vessby B, Uusitupa M, Hermansen K, Meyer BJ, Riccardi G, Rivellese AA. 2007. Fish oil, insulin 147.

sensitivity, insulin secretion and glucose tolerance in healthy people: is there any eff ect of fi sh oil supplementa-

tion in relation to the type of background diet and habitual dietary intake of n-6 and n-3 fatty acids? Nutr Metab

Cardiovasc Dis 17: 572-580.

Faeh D, Minehira K, Schwarz JM, Periasamy R, Park S, Tappy L. 2005. Eff ect of fructose overfeeding and fi sh oil admi-148.

nistration on hepatic de novo lipogenesis and insulin sensitivity in healthy men. Diabetes 54: 1907-1913.

Ahren B, Mari A, Fyfe CL, Tsofl iou F, Sneddon AA, Wahle KW, Winzell MS, Pacini G, Williams LM. 2009. Eff ects of 149.

conjugated linoleic acid plus n-3 polyunsaturated fatty acids on insulin secretion and estimated insulin sensitivity

in men. Eur J Clin Nutr 63: 778-786.

Hartweg J, Perera R, Montori V, Dinneen S, Neil HA, Farmer A. 2008. Omega-3 polyunsaturated fatty acids (PUFA) 150.

for type 2 diabetes mellitus. Cochrane Database Syst Rev CD003205.

Jump DB, Botolin D, Wang Y, Xu J, Christian B, Demeure O. 2005. Fatty acid regulation of hepatic gene transcrip-151.

tion. J Nutr 135: 2503-2506.

Zhao A, Yu J, Lew JL, Huang L, Wright SD, Cui J. 2004. Polyunsaturated fatty acids are FXR ligands and diff erentially 152.

regulate expression of FXR targets. DNA Cell Biol 23: 519-526.

Koves TR, Ussher JR, Noland RC, Slentz D, Mosedale M, Ilkayeva O, Bain J, Stevens R, Dyck JR, Newgard CB, Lopa-153.

schuk GD, Muoio DM. 2008. Mitochondrial overload and incomplete fatty acid oxidation contribute to skeletal

muscle insulin resistance. Cell Metab 7: 45-56.

Turner N, Bruce CR, Beale SM, Hoehn KL, So T, Rolph MS, Cooney GJ. 2007. Excess lipid availability increases mi-154.

tochondrial fatty acid oxidative capacity in muscle: evidence against a role for reduced fatty acid oxidation in

lipid-induced insulin resistance in rodents. Diabetes 56: 2085-2092.

1 3 8 References

Hancock CR, Han DH, Chen M, Terada S, Yasuda T, Wright DC, Holloszy JO. 2008. High-fat diets cause insulin resi-155.

stance despite an increase in muscle mitochondria. Proc Natl Acad Sci U S A 105: 7815-7820.

Ibrahimi A, Bonen A, Blinn WD, Hajri T, Li X, Zhong K, Cameron R, Abumrad NA. 1999. Muscle-specifi c overexpres-156.

sion of FAT/CD36 enhances fatty acid oxidation by contracting muscle, reduces plasma triglycerides and fatty

acids, and increases plasma glucose and insulin. J Biol Chem 274: 26761-26766.

Kraegen EW, Cooney GJ, Turner N. 2008. Muscle insulin resistance: a case of fat overconsumption, not mitochon-157.

drial dysfunction. Proc Natl Acad Sci U S A 105: 7627-7628.

Finck BN, Bernal-Mizrachi C, Han DH, Coleman T, Sambandam N, LaRiviere LL, Holloszy JO, Semenkovich CF, Kelly 158.

DP. 2005. A potential link between muscle peroxisome proliferator- activated receptor-alpha signaling and obesi-

ty-related diabetes. Cell Metab 1: 133-144.

Guerre-Millo M, Rouault C, Poulain P, Andre J, Poitout V, Peters JM, Gonzalez FJ, Fruchart JC, Reach G, Staels B. 2001. 159.

PPAR-alpha-null mice are protected from high-fat diet-induced insulin resistance. Diabetes 50: 2809-2814.

McLean JA, Tobin G (1987) Animal and human calorimetry. Cambridge [etc.]: Cambridge University Press. xiii, 338 160.

p.

Telting D, van der Zon GC, Dorrestijn J, Maassen JA. 2001. IRS-1 tyrosine phosphorylation refl ects insulin-induced 161.

metabolic and mitogenic responses in 3T3-L1 pre-adipocytes. Arch Physiol Biochem 109: 52-62.

Coenen KR, Gruen ML, Lee-Young RS, Puglisi MJ, Wasserman DH, Hasty AH. 2009. Impact of macrophage toll-like 162.

receptor 4 defi ciency on macrophage infi ltration into adipose tissue and the artery wall in mice. Diabetologia 52:

318-328.

Cunnane SC, McAdoo KR, Horrobin DF. 1986. n-3 Essential fatty acids decrease weight gain in genetically obese 163.

mice. Br J Nutr 56: 87-95.

Maeda H, Hosokawa M, Sashima T, Miyashita K. 2007. Dietary combination of fucoxanthin and fi sh oil attenuates 164.

the weight gain of white adipose tissue and decreases blood glucose in obese/diabetic KK-Ay mice. J Agric Food

Chem 55: 7701-7706.

Huber J, Loffl er M, Bilban M, Reimers M, Kadl A, Todoric J, Zeyda M, Geyeregger R, Schreiner M, Weichhart T, Lei-165.

tinger N, Waldhausl W, Stulnig TM. 2007. Prevention of high-fat diet-induced adipose tissue remodeling in obese

diabetic mice by n-3 polyunsaturated fatty acids. Int J Obes (Lond) 31: 1004-1013.

Ruzickova J, Rossmeisl M, Prazak T, Flachs P, Sponarova J, Veck M, Tvrzicka E, Bryhn M, Kopecky J. 2004. Omega-3 166.

PUFA of marine origin limit diet-induced obesity in mice by reducing cellularity of adipose tissue. Lipids 39: 1177-

1185.

Mori T, Kondo H, Hase T, Tokimitsu I, Murase T. 2007. Dietary fi sh oil upregulates intestinal lipid metabolism and 167.

reduces body weight gain in C57BL/6J mice. J Nutr 137: 2629-2634.

Ikemoto S, Takahashi M, Tsunoda N, Maruyama K, Itakura H, Ezaki O. 1996. High-fat diet-induced hyperglycemia 168.

and obesity in mice: diff erential eff ects of dietary oils. Metabolism 45: 1539-1546.

Madsen L, Pedersen LM, Liaset B, Ma T, Petersen RK, van den BS, Pan J, Muller-Decker K, Dulsner ED, Kleemann R, 169.

Kooistra T, Doskeland SO, Kristiansen K. 2008. cAMP-dependent signaling regulates the adipogenic eff ect of n-6

polyunsaturated fatty acids. J Biol Chem 283: 7196-7205.

Bailey CJ, Atkins TW, Conner MJ, Manley CG, Matty AJ. 1975. Diurnal variations of food consumption, plasma glu-170.

cose and plasma insulin concentrations in lean and obese hyperglycaemic mice. Horm Res 6: 380-386.

Kohsaka A, Laposky AD, Ramsey KM, Estrada C, Joshu C, Kobayashi Y, Turek FW, Bass J. 2007. High-fat diet disrupts 171.

behavioral and molecular circadian rhythms in mice. Cell Metab 6: 414-421.

Corpeleijn E, Mensink M, Kooi ME, Roekaerts PM, Saris WH, Blaak EE. 2008. Impaired skeletal muscle substrate 172.

oxidation in glucose-intolerant men improves after weight loss. Obesity (Silver Spring) 16: 1025-1032.

Kelley DE, Goodpaster B, Wing RR, Simoneau JA. 1999. Skeletal muscle fatty acid metabolism in association with 173.

insulin resistance, obesity, and weight loss. Am J Physiol 277: E1130-E1141.

1 3 9References

Galgani JE, Heilbronn LK, Azuma K, Kelley DE, Albu JB, Pi-Sunyer X, Smith SR, Ravussin E. 2008. Metabolic fl exibility 174.

in response to glucose is not impaired in people with type 2 diabetes after controlling for glucose disposal rate.

Diabetes 57: 841-845.

DeFronzo RA, Jacot E, Jequier E, Maeder E, Wahren J, Felber JP. 1981. The eff ect of insulin on the disposal of intra-175.

venous glucose. Results from indirect calorimetry and hepatic and femoral venous catheterization. Diabetes 30:

1000-1007.

Neschen S, Morino K, Dong J, Wang-Fischer Y, Cline GW, Romanelli AJ, Rossbacher JC, Moore IK, Regittnig W, Mu-176.

noz DS, Kim JH, Shulman GI. 2007. n-3 Fatty Acids Preserve Insulin Sensitivity In Vivo in a Peroxisome Proliferator-

Activated Receptor-{alpha}-Dependent Manner. Diabetes 56: 1034-1041.

Clee SM, Attie AD. 2007. The genetic landscape of type 2 diabetes in mice. Endocr Rev 28: 48-83.177.

Ayala JE, Bracy DP, McGuinness OP, Wasserman DH. 2006. Considerations in the design of hyperinsulinemic-eug-178.

lycemic clamps in the conscious mouse. Diabetes 55: 390-397.

Postic C, Girard J. 2008. Contribution of 179. de novo fatty acid synthesis to hepatic steatosis and insulin resistance:

lessons from genetically engineered mice. J Clin Invest 118: 829-838.

Tamura S, Shimomura I. 2005. Contribution of adipose tissue and 180. de novo lipogenesis to nonalcoholic fatty liver

disease. J Clin Invest 115: 1139-1142.

Gentile CL, Pagliassotti MJ. 2008. The role of fatty acids in the development and progression of nonalcoholic fatty 181.

liver disease. J Nutr Biochem 19: 567-576.

Westerbacka J, Lammi K, Hakkinen AM, Rissanen A, Salminen I, Aro A, Yki-Jarvinen H. 2005. Dietary fat content 182.

modifi es liver fat in overweight nondiabetic subjects. J Clin Endocrinol Metab 90: 2804-2809.

van der Meer RW, Hammer S, Lamb HJ, Frolich M, Diamant M, Rijzewijk LJ, de RA, Romijn JA, Smit JW. 2008. Eff ects 183.

of short-term high-fat, high-energy diet on hepatic and myocardial triglyceride content in healthy men. J Clin

Endocrinol Metab 93: 2702-2708.

Lombardo YB, Chicco AG. 2006. Eff ects of dietary polyunsaturated n-3 fatty acids on dyslipidemia and insulin 184.

resistance in rodents and humans. A review. J Nutr Biochem 17: 1-13.

Lin J, Yang R, Tarr PT, Wu PH, Handschin C, Li S, Yang W, Pei L, Uldry M, Tontonoz P, Newgard CB, Spiegelman BM. 185.

2005. Hyperlipidemic eff ects of dietary saturated fats mediated through PGC-1beta coactivation of SREBP. Cell

120: 261-273.

Mao J, DeMayo FJ, Li H, bu-Elheiga L, Gu Z, Shaikenov TE, Kordari P, Chirala SS, Heird WC, Wakil SJ. 2006. Liver-spe-186.

cifi c deletion of acetyl-CoA carboxylase 1 reduces hepatic triglyceride accumulation without aff ecting glucose

homeostasis. Proc Natl Acad Sci U S A 103: 8552-8557.

Savage DB, Choi CS, Samuel VT, Liu ZX, Zhang D, Wang A, Zhang XM, Cline GW, Yu XX, Geisler JG, Bhanot S, Monia 187.

BP, Shulman GI. 2006. Reversal of diet-induced hepatic steatosis and hepatic insulin resistance by antisense oligo-

nucleotide inhibitors of acetyl-CoA carboxylases 1 and 2. J Clin Invest 116: 817-824.

Gutierrez-Juarez R, Pocai A, Mulas C, Ono H, Bhanot S, Monia BP, Rossetti L. 2006. Critical role of stearoyl-CoA 188.

desaturase-1 (SCD1) in the onset of diet-induced hepatic insulin resistance. J Clin Invest 116: 1686-1695.

Jump DB. 2008. N-3 polyunsaturated fatty acid regulation of hepatic gene transcription. Curr Opin Lipidol 19: 189.

242-247.

Shimano H, Horton JD, Shimomura I, Hammer RE, Brown MS, Goldstein JL. 1997. Isoform 1c of sterol regulatory 190.

element binding protein is less active than isoform 1a in livers of transgenic mice and in cultured cells. J Clin

Invest 99: 846-854.

Plosch T, Kok T, Bloks VW, Smit MJ, Havinga R, Chimini G, Groen AK, Kuipers F. 2002. Increased hepatobiliary and 191.

fecal cholesterol excretion upon activation of the liver X receptor is independent of ABCA1. J Biol Chem 277:

33870-33877.

Gautier T, Tietge UJF, Boverhof R, Perton FG, Le GN, Masson D, Rensen PC, Havekes LM, Lagrost L, Kuipers F. 2007. 192.

Hepatic lipid accumulation in apolipoprotein C-I-defi cient mice is potentiated by cholesteryl ester transfer pro-

tein. J Lipid Res 48: 30-40.

1 4 0 References

Fraser R. 1970. Size and lipid composition of chylomicrons of diff erent Svedberg units of fl otation. J Lipid Res 11: 193.

60-65.

Zhang Y, Agarwal KC, Beylot M, Soloviev MV, David F, Reider MW, Anderson VE, Tserng KY, Brunengraber H. 1994. 194.

Nonhomogeneous labeling of liver extra-mitochondrial acetyl-CoA. Implications for the probing of lipogenic

acetyl-CoA via drug acetylation and for the production of acetate by the liver. J Biol Chem 269: 11025-11029.

Bederman IR, Reszko AE, Kasumov T, David F, Wasserman DH, Kelleher JK, Brunengraber H. 2004. Zonation of labe-195.

ling of lipogenic acetyl-CoA across the liver: implications for studies of lipogenesis by mass isotopomer analysis.

J Biol Chem 279: 43207-43216.

Hellerstein MK, Wu K, Kaempfer S, Kletke C, Shackleton CH. 1991. Sampling the lipogenic hepatic acetyl-CoA pool 196.

in vivo in the rat. Comparison of xenobiotic probe to values predicted from isotopomeric distribution in circula-

ting lipids and measurement of lipogenesis and acetyl-CoA dilution. J Biol Chem 266: 10912-10919.

Wiegman CH, Bandsma RH, Ouwens M, van der Sluijs FH, Havinga R, Boer T, Reijngoud DJ, Romijn JA, Kuipers F. 197.

2003. Hepatic VLDL production in ob/ob mice is not stimulated by massive de novo lipogenesis but is less sensi-

tive to the suppressive eff ects of insulin. Diabetes 52: 1081-1089.

Bandsma RH, Wiegman CH, Herling AW, Burger HJ, ter HA, Meijer AJ, Romijn JA, Reijngoud DJ, Kuipers F. 2001. 198.

Acute inhibition of glucose-6-phosphate translocator activity leads to increased de novo lipogenesis and deve-

lopment of hepatic steatosis without aff ecting VLDL production in rats. Diabetes 50: 2591-2597.

Jung HR, Turner SM, Neese RA, Young SG, Hellerstein MK. 1999. Metabolic adaptations to dietary fat malabsorp-199.

tion in chylomicron-defi cient mice. Biochem J 343 Pt 2: 473-478.

Delgado TC, Pinheiro D, Caldeira M, Castro MM, Geraldes CF, Lopez-Larrubia P, Cerdan S, Jones JG. 2009. Sources of 200.

hepatic triglyceride accumulation during high-fat feeding in the healthy rat. NMR Biomed 22: 310-317.

Schwarz JM, Neese RA, Turner S, Dare D, Hellerstein MK. 1995. Short-term alterations in carbohydrate energy 201.

intake in humans. Striking eff ects on hepatic glucose production, de novo lipogenesis, lipolysis, and whole-body

fuel selection. J Clin Invest 96: 2735-2743.

Hudgins LC, Hellerstein M, Seidman C, Neese R, Diakun J, Hirsch J. 1996. Human fatty acid synthesis is stimulated 202.

by a eucaloric low fat, high carbohydrate diet. J Clin Invest 97: 2081-2091.

Bassilian S, Ahmed S, Lim SK, Boros LG, Mao CS, Lee WN. 2002. Loss of regulation of lipogenesis in the Zucker dia-203.

betic rat. II. Changes in stearate and oleate synthesis. Am J Physiol Endocrinol Metab 282: E507-E513.

Doege H, Grimm D, Falcon A, Tsang B, Storm TA, Xu H, Ortegon AM, Kazantzis M, Kay MA, Stahl A. 2008. Silencing 204.

of hepatic fatty acid transporter protein 5 in vivo reverses diet-induced non-alcoholic fatty liver disease and im-

proves hyperglycemia. J Biol Chem 283: 22186-22192.

Cheema SK, Agellon LB. 1999. Metabolism of cholesterol is altered in the liver of C3H mice fed fats enriched with 205.

diff erent C-18 fatty acids. J Nutr 129: 1718-1724.

Rumsey SC, Galeano NF, Lipschitz B, Deckelbaum RJ. 1995. Oleate and other long chain fatty acids stimulate low 206.

density lipoprotein receptor activity by enhancing acyl coenzyme A:cholesterol acyltransferase activity and alte-

ring intracellular regulatory cholesterol pools in cultured cells. J Biol Chem 270: 10008-10016.

Brown MS, Goldstein JL. 1997. The SREBP pathway: regulation of cholesterol metabolism by proteolysis of a mem-207.

brane-bound transcription factor. Cell 89: 331-340.

Goldstein JL, Rawson RB, Brown MS. 2002. Mutant mammalian cells as tools to delineate the sterol regulatory ele-208.

ment-binding protein pathway for feedback regulation of lipid synthesis. Arch Biochem Biophys 397: 139-148.

Park Y, Harris WS. 2003. Omega-3 fatty acid supplementation accelerates chylomicron triglyceride clearance. J 209.

Lipid Res 44: 455-463.

Qi K, Fan C, Jiang J, Zhu H, Jiao H, Meng Q, Deckelbaum RJ. 2008. Omega-3 fatty acid containing diets decrease 210.

plasma triglyceride concentrations in mice by reducing endogenous triglyceride synthesis and enhancing the

blood clearance of triglyceride-rich particles. Clin Nutr 27: 424-430.

1 4 1References

Murphy MC, Zampelas A, Puddicombe SM, Furlonger NP, Morgan LM, Williams CM. 1993. Pretranslational regu-211.

lation of the expression of the lipoprotein lipase (EC 3.1.1.34) gene by dietary fatty acids in the rat. Br J Nutr 70:

727-736.

Benhizia F, Hainault I, Serougne C, Lagrange D, Hajduch E, Guichard C, Malewiak MI, Quignard-Boulange A, Lavau 212.

M, Griglio S. 1994. Eff ects of a fi sh oil-lard diet on rat plasma lipoproteins, liver FAS, and lipolytic enzymes. Am J

Physiol 267: E975-E982.

Gibbons GF, Bartlett SM, Sparks CE, Sparks JD. 1992. Extracellular fatty acids are not utilized directly for the synthe-213.

sis of very-low-density lipoprotein in primary cultures of rat hepatocytes. Biochem J 287 ( Pt 3): 749-753.

Gibbons GF, Wiggins D, Brown AM, Hebbachi AM. 2004. Synthesis and function of hepatic very-low-density li-214.

poprotein. Biochem Soc Trans 32: 59-64.

Alwayn IP, Andersson C, Zauscher B, Gura K, Nose V, Puder M. 2005. Omega-3 fatty acids improve hepatic steatosis 215.

in a murine model: potential implications for the marginal steatotic liver donor. Transplantation 79: 606-608.

Lin Y, Smit MJ, Havinga R, Verkade HJ, Vonk RJ, Kuipers F. 1995. Diff erential eff ects of eicosapentaenoic acid on gly-216.

cerolipid and apolipoprotein B metabolism in primary human hepatocytes compared to HepG2 cells and primary

rat hepatocytes. Biochim Biophys Acta 1256: 88-96.

Pan M, Cederbaum AI, Zhang YL, Ginsberg HN, Williams KJ, Fisher EA. 2004. Lipid peroxidation and oxidant stress 217.

regulate hepatic apolipoprotein B degradation and VLDL production. J Clin Invest 113: 1277-1287.

Repa JJ, Mangelsdorf DJ. 2002. The liver X receptor gene team: potential new players in atherosclerosis. Nat Med 218.

8: 1243-1248.

Ross SE, Erickson RL, Gerin I, DeRose PM, Bajnok L, Longo KA, Misek DE, Kuick R, Hanash SM, Atkins KB, Andresen 219.

SM, Nebb HI, Madsen L, Kristiansen K, Macdougald OA. 2002. Microarray analyses during adipogenesis: understan-

ding the eff ects of Wnt signaling on adipogenesis and the roles of liver X receptor alpha in adipocyte metabolism.

Mol Cell Biol 22: 5989-5999.

Ide T, Shimano H, Yoshikawa T, Yahagi N, Amemiya-Kudo M, Matsuzaka T, Nakakuki M, Yatoh S, Iizuka Y, Tomita S, 220.

Ohashi K, Takahashi A, Sone H, Gotoda T, Osuga J, Ishibashi S, Yamada N. 2003. Cross-talk between peroxisome

proliferator-activated receptor (PPAR) alpha and liver X receptor (LXR) in nutritional regulation of fatty acid meta-

bolism. II. LXRs suppress lipid degradation gene promoters through inhibition of PPAR signaling. Mol Endocrinol

17: 1255-1267.

Makishima M, Okamoto AY, Repa JJ, Tu H, Learned RM, Luk A, Hull MV, Lustig KD, Mangelsdorf DJ, Shan B. 1999. 221.

Identifi cation of a nuclear receptor for bile acids. Science 284: 1362-1365.

Horton JD, Bashmakov Y, Shimomura I, Shimano H. 1998. Regulation of sterol regulatory element binding proteins 222.

in livers of fasted and refed mice. Proc Natl Acad Sci U S A 95: 5987-5992.

Matsuda M, Korn BS, Hammer RE, Moon YA, Komuro R, Horton JD, Goldstein JL, Brown MS, Shimomura I. 2001. 223.

SREBP cleavage-activating protein (SCAP) is required for increased lipid synthesis in liver induced by cholesterol

deprivation and insulin elevation. Genes Dev 15: 1206-1216.

Ferrannini E, Barrett EJ, Bevilacqua S, DeFronzo RA. 1983. Eff ect of fatty acids on glucose production and utilization 224.

in man. J Clin Invest 72: 1737-1747.

Kelley DE, Mokan M, Simoneau JA, Mandarino LJ. 1993. Interaction between glucose and free fatty acid metabo-225.

lism in human skeletal muscle. J Clin Invest 92: 91-98.

Boden G, Chen X, Ruiz J, White JV, Rossetti L. 1994. Mechanisms of fatty acid-induced inhibition of glucose uptake. 226.

J Clin Invest 93: 2438-2446.

Yahagi N, Shimano H, Hasty AH, Amemiya-Kudo M, Okazaki H, Tamura Y, Iizuka Y, Shionoiri F, Ohashi K, Osuga J, Ha-227.

rada K, Gotoda T, Nagai R, Ishibashi S, Yamada N. 1999. A crucial role of sterol regulatory element-binding protein-1

in the regulation of lipogenic gene expression by polyunsaturated fatty acids. J Biol Chem 274: 35840-35844.

Petit V, Arnould L, Martin P, Monnot MC, Pineau T, Besnard P, Niot I. 2007. Chronic high-fat diet aff ects intestinal fat 228.

absorption and postprandial triglyceride levels in the mouse. J Lipid Res 48: 278-287.

1 4 2 References

Tan CY, Vidal-Puig A. 2008. Adipose tissue expandability: the metabolic problems of obesity may arise from the 229.

inability to become more obese. Biochem Soc Trans 36: 935-940.

Medina-Gomez G, Gray SL, Yetukuri L, Shimomura K, Virtue S, Campbell M, Curtis RK, Jimenez-Linan M, Blount M, 230.

Yeo GS, Lopez M, Seppanen-Laakso T, Ashcroft FM, Oresic M, Vidal-Puig A. 2007. PPAR gamma 2 prevents lipotoxi-

city by controlling adipose tissue expandability and peripheral lipid metabolism. PLoS Genet 3: e64.

Kim JY, van de WE, Laplante M, Azzara A, Trujillo ME, Hofmann SM, Schraw T, Durand JL, Li H, Li G, Jelicks LA, Mehler 231.

MF, Hui DY, Deshaies Y, Shulman GI, Schwartz GJ, Scherer PE. 2007. Obesity-associated improvements in metabolic

profi le through expansion of adipose tissue. J Clin Invest 117: 2621-2637.

Miyazaki Y, Mahankali A, Matsuda M, Glass L, Mahankali S, Ferrannini E, Cusi K, Mandarino LJ, DeFronzo RA. 2001. 232.

Improved glycemic control and enhanced insulin sensitivity in type 2 diabetic subjects treated with pioglitazone.

Diabetes Care 24: 710-719.

Simha V, Garg A. 2006. Lipodystrophy: lessons in lipid and energy metabolism. Curr Opin Lipidol 17: 162-169.233.

Gavrilova O, Marcus-Samuels B, Graham D, Kim JK, Shulman GI, Castle AL, Vinson C, Eckhaus M, Reitman ML. 2000. 234.

Surgical implantation of adipose tissue reverses diabetes in lipoatrophic mice. J Clin Invest 105: 271-278.

Galgani JE, Moro C, Ravussin E. 2008. Metabolic fl exibility and insulin resistance. Am J Physiol Endocrinol Metab 235.

295: E1009-E1017.

Bell KS, Schmitz-Peiff er C, Lim-Fraser M, Biden TJ, Cooney GJ, Kraegen EW. 2000. Acute reversal of lipid-induced 236.

muscle insulin resistance is associated with rapid alteration in PKC-theta localization. Am J Physiol Endocrinol

Metab 279: E1196-E1201.

Farese RV, Sajan MP, Standaert ML. 2005. Atypical protein kinase C in insulin action and insulin resistance. Biochem 237.

Soc Trans 33: 350-353.

Pajari AM, Mutanen M. 1999. Phospholipid fatty acid composition and protein kinase C activity in the large intes-238.

tine of rats fed on butter and coconut-oil diets. Br J Nutr 82: 411-418.

Savage DB, Petersen KF, Shulman GI. 2007. Disordered lipid metabolism and the pathogenesis of insulin resi-239.

stance. Physiol Rev 87: 507-520.

Oakes ND, Bell KS, Furler SM, Camilleri S, Saha AK, Ruderman NB, Chisholm DJ, Kraegen EW. 1997. Diet-induced 240.

muscle insulin resistance in rats is ameliorated by acute dietary lipid withdrawal or a single bout of exercise: pa-

rallel relationship between insulin stimulation of glucose uptake and suppression of long-chain fatty acyl-CoA.

Diabetes 46: 2022-2028.

Donnelly KL, Smith CI, Schwarzenberg SJ, Jessurun J, Boldt MD, Parks EJ. 2005. Sources of fatty acids stored in liver 241.

and secreted via lipoproteins in patients with nonalcoholic fatty liver disease. J Clin Invest 115: 1343-1351.

Donnelly KL, Margosian MR, Sheth SS, Lusis AJ, Parks EJ. 2004. Increased lipogenesis and fatty acid reesterifi cation 242.

contribute to hepatic triacylglycerol stores in hyperlipidemic Txnip-/- mice. J Nutr 134: 1475-1480.

Rafi q N, Younossi ZM. 2008. Eff ects of weight loss on nonalcoholic fatty liver disease. Semin Liver Dis 28: 427-433.243.

Ruge T, Hodson L, Cheeseman J, Dennis AL, Fielding BA, Humphreys SM, Frayn KN, Karpe F. 2009. Fasted to fed 244.

traffi cking of Fatty acids in human adipose tissue reveals a novel regulatory step for enhanced fat storage. J Clin

Endocrinol Metab 94: 1781-1788.

Belfort R, Harrison SA, Brown K, Darland C, Finch J, Hardies J, Balas B, Gastaldelli A, Tio F, Pulcini J, Berria R, Ma JZ, 245.

Dwivedi S, Havranek R, Fincke C, DeFronzo R, Bannayan GA, Schenker S, Cusi K. 2006. A placebo-controlled trial of

pioglitazone in subjects with nonalcoholic steatohepatitis. N Engl J Med 355: 2297-2307.

Lutchman G, Modi A, Kleiner DE, Promrat K, Heller T, Ghany M, Borg B, Loomba R, Liang TJ, Premkumar A, Hoofna-246.

gle JH. 2007. The eff ects of discontinuing pioglitazone in patients with nonalcoholic steatohepatitis. Hepatology

46: 424-429.

Dulloo AG, Gubler M, Montani JP, Seydoux J, Solinas G. 2004. Substrate cycling between 247. de novo lipogenesis and

lipid oxidation: a thermogenic mechanism against skeletal muscle lipotoxicity and glucolipotoxicity. Int J Obes

Relat Metab Disord 28 Suppl 4: S29-S37.

1 4 3References

Kotronen A, Seppanen-Laakso T, Westerbacka J, Kiviluoto T, Arola J, Ruskeepaa AL, Oresic M, Yki-Jarvinen H. 2009. 248.

Hepatic stearoyl-CoA desaturase (SCD)-1 activity and diacylglycerol but not ceramide concentrations are incre-

ased in the nonalcoholic human fatty liver. Diabetes 58: 203-208.

Peter A, Weigert C, Staiger H, Rittig K, Cegan A, Lutz P, Machicao F, Haring HU, Schleicher E. 2008. Induction of 249.

stearoyl-CoA desaturase protects human arterial endothelial cells against lipotoxicity. Am J Physiol Endocrinol

Metab 295: E339-E349.

Zhang L, Ge L, Tran T, Stenn K, Prouty SM. 2001. Isolation and characterization of the human stearoyl-CoA desatu-250.

rase gene promoter: requirement of a conserved CCAAT cis-element. Biochem J 357: 183-193.

Vidal-Puig A, Jimenez-Linan M, Lowell BB, Hamann A, Hu E, Spiegelman B, Flier JS, Moller DE. 1996. Regulation of 251.

PPAR gamma gene expression by nutrition and obesity in rodents. J Clin Invest 97: 2553-2561.

Inoue M, Ohtake T, Motomura W, Takahashi N, Hosoki Y, Miyoshi S, Suzuki Y, Saito H, Kohgo Y, Okumura T. 2005. 252.

Increased expression of PPARgamma in high fat diet-induced liver steatosis in mice. Biochem Biophys Res Com-

mun 336: 215-222.

Parks EJ, Hellerstein MK. 2006. Thematic review series: patient-oriented research. Recent advances in liver triacyl-253.

glycerol and fatty acid metabolism using stable isotope labeling techniques. J Lipid Res 47: 1651-1660.

Hornberg JJ, Bruggeman FJ, Bakker BM, Westerhoff HV. 2007. Metabolic control analysis to identify optimal drug 254.

targets. Prog Drug Res 64: 171, 173-171, 189.

Hellerstein MK. 2008. Exploiting complexity and the robustness of network architecture for drug discovery. J 255.

Pharmacol Exp Ther 325: 1-9.

1 4 5Summary

Summary

The body possesses sensor systems that respond to changes in nutrient supply, thereby enabling

adequate adaptation of metabolic processes to nutrient availability. Superfl uous nutrients are stored

if energy supply exceeds energy consumption. These stores are in turn used when energy supply is

limited. Such metabolic fl exibility ensures proper functioning of living organisms under changing

conditions.

Adaptive metabolic responses occur via modifi cations of metabolic fl uxes. A fl ux represents the

fl ow of molecules through a series of chemical reactions that constitute a biochemical pathway. The

fl ow rate is determined by nutrient availability and enzyme activities. Enzyme activity per se may

also be altered by nutrient availability. Thus, nutrient status determines metabolic fl ux by both direct

and indirect mechanisms.

The presence and activity of metabolic enzymes are tightly controlled. Enzyme synthesis is initi-

ated by the transcription of a specifi c genetic code from DNA. The enzyme is synthesized upon

translation of this code. The transcription process is partly dependent on DNA-binding of specifi c

transcription factors. Some transcription factors (referred to as nuclear receptors) are activated upon

binding of ligands, and subsequently promote or inhibit transcription of their target genes. In recent

years, nutrients have emerged as ligands for a select group of nuclear receptors. These represent nu-

trient sensors that are able to adapt metabolic enzyme transcription in response to changes in nutri-

ent status. Thus, metabolic fl uxes can be attenuated by alterations in transcription factor activity.

As stated earlier, metabolic fl exibility enables short-term adaptation to acute changes in nutrient

availability. However, metabolic fl uxes will be persistently modifi ed upon chronic energy oversupply

and nutritional dysbalance. These adaptive responses may in the long term predispose to the devel-

opment of metabolic abnormalities such as obesity, hepatic steatosis and type 2 diabetes. Therefore,

it is important to gain insight into the adaptive modulations of metabolic fl uxes. Furthermore, the

possibility to attenuate metabolic disturbances via enzyme transcription has sparked the interest

to design drugs that modulate transcription factor action. Current knowledge on the regulation of

metabolic fl uxes is limited. Deeper insights into these regulatory pathways will contribute to the

development of new drugs.

The studies described in this thesis consider physiological adaptations that occur in response to

changes in nutrient availability. In particular, the role of specifi c transcription factors was addressed.

In Chapter 2, we determined the role of the ‘Liver X Receptor’(LXR) in the liver during the feeding-

to-fasting transition. The two LXR isotypes α en β are both involved in the regulation of cholesterol

and fatty acid metabolism. LXR activity is determined by cellular cholesterol content. Furthermore,

glucose has been postulated to serve as a LXR ligand. We therefore challenged mice with a glucose-

rich diet. Surprisingly, we did not observe an induction of direct LXR target genes upon glucose

exposure. Furthermore, no diff erences in gene expression were observed between normal (‘wild-

type’) mice and mice in which LXRα action was abolished (‘LXRα knockouts’). However, when these

animals were fasted, hepatic glycogen depletion was found to be delayed in LXRα knockout mice.

Furthermore, major fl uxes involved in hepatic glucose metabolism were found to be reduced in

these animals. We also observed that fasting-induced hepatic steatosis was diminished in LXRα

1 4 6 Summary

knockouts. Therefore, LXRα appears to be required for an adequate attenuation of metabolic fl uxes

under conditions of low nutrient supply, when the body starts using its reserves.

The ‘Farnesoid X Receptor’ (FXR) is a transcription factor that is activated by bile acids. Upon fast-

ing, FXR knockouts exhibit an impaired ability to maintain hepatic glucose metabolism. The results

of the studies described in Chapter 3 indicate that FXR exerts a regulatory role in intestinal glucose

metabolism. The increase in blood glucose concentrations upon an oral glucose challenge was

found to be diminished in FXR knockouts. Additional studies revealed that intestinal glucose uptake

was reduced in these mice. This is explained by an increased glucose fl ux trough an alternative

route. From these studies we conclude that FXR inactivation induces changes in intestinal glucose

metabolism.

‘Peroxisome Proliferator Activated Receptors’ (PPARs) represent a group of transcription factors

that is activated by fatty acids. PPARα is an important regulator of hepatic lipid metabolism, and its

action has been shown to be crucial for the adaptations that occur in response to fasting. These in-

clude the induction of fatty acid oxidation while limiting glucose consumption. PPARα furthermore

induces a number of systems that protect against damage by fatty acid oxidation products. The

studies in Chapter 4 add to the current understanding of the adaptations that occur in response to

an increased PPARα activity in the liver. Mice were treated with a pharmacological PPARα agonist.

The hepatic expression of fatty acid oxidation genes was consequently increased. To our surprise, we

also observed an induction of genes involved in hepatic fatty acid synthesis (‘lipogenesis’). This in-

duction was found to depend on the presence of the transcription factor ‘Sterol Regulatory Element

Binding Protein 1c’ (SREBP-1c). These transcriptional changes were translated into an increase in the

lipogenic fl ux, and were furthermore paralleled by an increased hepatic lipid content. We also found

that glucose fl uxes were altered. Altogether, our data support the co-existence of hepatic fatty acid

oxidation and lipogenesis. The induction of hepatic fatty acid synthesis and the accumulation of

lipid in the liver may represent a physiological mechanism by which the liver is protected against

damage by fatty acids and their oxidation products.

Chapter 5 and 6 address the metabolic consequences of a increased dietary fat supply. We stud-

ied the eff ects of two diff erent high-fat diets. The fi rst diet was based on beef tallow and therefore

rich in saturated fatty acids. In the second diet, the beef tallow was partially replaced by fi sh oil, rich

in n-3 polyunsaturated fatty acids (PUFA). Intake of PUFA reduces atherosclerotic and cardiovascu-

lar risk in humans. Mice were fed either of the two high-fat diets during six weeks. The increased

dietary fat intake resulted in an increase in whole-body fat oxidation as compared to mice receiv-

ing a regular (low-fat) diet. This eff ect was most pronounced in mice fed the fi sh oil-enriched diet

(Chapter 5). Both high-fat diets induced adiposity, because energy consumption was increased,

while energy expenditure remained unaltered (Chapter 5). Mice fed the tallow-rich diet exhibited

an increased lipogenic fl ux in their livers. This was paralleled by an increased expression of lipogenic

genes (Chapter 6). The amount of lipid secreted by the liver was however comparable to mice fed

the low-fat diet. As a consequence, the net storage of hepatic lipid was increased in tallow-fed mice.

Partial substitution of the saturated fat by fi sh oil resulted in a suppression of both lipogenic gene ex-

pression and the fl ux through this pathway, thereby protecting against hepatic lipid accumulation.

Therefore, fi sh oil exerts benefi cial eff ect on lipid metabolism, which contribute to the prevention of

hepatic steatosis. We fi nally evaluated the consequences for glucose fl uxes. Intake of the tallow-rich

1 4 7Summary

diet induced insulin resistance of glucose metabolism. This predisposes to development of type 2

diabetes in the long term. Chronic oversupply of dietary fat resulted in a persistent reliance on fat

oxidation, which resulted in a reduction of glucose uptake and oxidation. In contrast to the reported

improvements in lipid metabolism, fi sh oil substitution did not rescue glycemia: it even potentiated

the development of insulin resistance in these high fat-fed mice (Chapter 5). A protective eff ect of

fi sh oil on the development of insulin resistance has been observed in previous studies. Therefore,

additional studies are required to evaluate the eff ects of fi sh oil under diff erent conditions.

These studies add to the current understanding on the action of transcription factors that control

metabolic fl uxes. An increased activity of specifi c transcription factors appears to be required to

ensure proper handling of fatty acids, in order to limit hepatic damage. However, in most cases, this

is accompanied by hepatic lipid accumulation. Although hepatic steatosis per se is relatively harm-

less, lipid accumulation in the liver may increase the risk to develop liver disease, in particular if it is

accompanied by an infl ammatory event. The studies furthermore show that inhibition of transcrip-

tion factor action can result in the re-arrangement of metabolic fl uxes. Such adaptations may be

diff erent from what is predicted from target gene expression patterns. In addition, the interactions

of enzymes and the net eff ect on global fl uxes may vary under diff erent metabolic conditions. The

metabolic consequences of transcription factor action should therefore always be evaluated under

relevant conditions. Finally, a change in the fl ux through a particular pathway may aff ect the fl ux

through another route. Thus, the application of fl uxomics in vivo is required to obtain a complete

picture on the eff ects of drugs designed to modulate transcription factor activity. Careful evaluation

of the global metabolic consequences will contribute to the development of future drugs.

1 4 9Samenvatting

Samenvatting

Het lichaam bevat sensoren die het aanbod van voedingsstoff en (‘nutriënten’) opmerken. Hierdoor

is het mogelijk de stofwisseling (het ‘metabolisme’) af te stemmen op de beschikbaarheid van voe-

dingsstoff en. Zolang er voldoende energie beschikbaar is, worden overtollige voedingstoff en opge-

slagen. Deze reserves kunnen worden aangesproken wanneer het aanbod niet kan voorzien in de

energiebehoefte. Door deze fl exibiliteit van het metabolisme is het lichaam in staat om te blijven

functioneren onder wisselende omstandigheden.

Aanpassingen in de stofwisseling vinden plaats door het veranderen van metabole fl uxen. Een

fl ux is de stroom van moleculen door een aantal opeenvolgende chemische reacties. Deze vor-

men samen een biochemisch reactiepad. De grootte van een metabole fl ux wordt bepaald door de

beschikbaarheid van voedingsstoff en, en de activiteit van katalysatoren die de chemische reacties

versnellen (‘enzymen’). Deze activiteit wordt mede bepaald door de aanwezigheid van voedings-

stoff en. De voedingsstatus heeft daarom een directe én indirecte invloed op de grootte van meta-

bole fl uxen.

Zowel de activiteit als de aanwezigheid van metabole enzymen wordt nauwkeurig gereguleerd.

De aanmaak van een enzym begint met het afl ezen van een specifi eke genetische code op het DNA

(‘transcriptie’). Deze code wordt vertaald, waarna het enzym geproduceerd kan worden. De voort-

gang van het transcriptieproces wordt onder meer bepaald door de aanwezigheid van zogenaam-

de transcriptiefactoren op het DNA. Sommige transcriptiefactoren, genaamd nucleaire receptoren,

worden geactiveerd door de binding van specifi eke moleculen (‘liganden’), en verhogen of verlagen

vervolgens de transcriptie van bepaalde genen (‘target genen’). In de afgelopen jaren is duidelijk

geworden dat voedingsstoff en als ligand dienen voor een select aantal nucleaire receptoren. Deze

vormen een groep ‘nutriënt sensoren’ die ervoor zorgen dat de transcriptie van metabole enzymen

wordt aangepast wanneer het aanbod van voedingsstoff en verandert. Door het aanpassen van de

activiteit van transcriptiefactoren kunnen metabole fl uxen dus gestuurd worden.

Zoals gezegd maakt de fl exibiliteit van het metabolisme korte-termijn aanpassingen aan wisse-

lende omstandigheden mogelijk. Wanneer het aanbod van energie en voedingsstoff en gedurende

langere tijd groot is, zullen metabole fl uxen blijvend veranderen. Hierdoor zal de kans op het ont-

staan van metabole verstoringen zoals overgewicht, leververvetting en suikerziekte toenemen. Het

is daarom belangrijk te achterhalen op welke manier de veranderingen in metabole fl uxen precies

tot stand komen. Bovendien biedt het veranderen van enzymtranscriptie de mogelijkheid om meta-

bole verstoringen bij te sturen. Er is dan ook een groeiende interesse om de activiteit van transcrip-

tiefactoren met medicijnen te beïnvloeden. De kennis over de regulering van metabole fl uxen door

transcriptiefactoren is op dit moment beperkt. Wanneer deze wordt uitgebreid kan de werking van

toekomstige medicijnen verbeterd worden.

In dit proefschrift zijn de metabole aanpassingen die optreden als gevolg van veranderingen in

de voedingsstatus gedeeltelijk in kaart gebracht. Hierbij is de rol van een aantal transcriptiefactoren

nader onderzocht. In Hoofdstuk 2 werd de functie van de ‘Liver X Receptor’ (LXR) in de lever be-

studeerd tijdens de overgang van voeden naar vasten. Er zijn 2 vormen van LXR, α en β. Van beide

vormen is bekend dat zij het metabolisme van cholesterol en vetzuren beïnvloeden. De activiteit

1 5 0 Samenvatting

van LXR wordt bepaald door de cholesterolconcentratie in de cel. Er is daarnaast beperkt bewijs dat

glucose als LXR ligand kan dienen. Om dit verder te bestuderen, boden we muizen een glucose rijk

dieet aan. De transcriptie van directe LXR target genen werd hierdoor echter niet verhoogd. Ook was

er geen verschil waarneembaar in deze transcriptie wanneer normale muizen werden vergeleken

met dieren waarin de werking van LXRα uitgeschakeld is (‘LXRα knockouts’). Wanneer beide soorten

muizen werden gevast, bleek dat de afbraak van glycogeen (de opslagvorm van glucose) vertraagd

was in de levers van LXRα knockouts. Bovendien waren de fl uxen die van belang zijn voor de omzet-

ting van zowel glucose als glycogeen verlaagd in deze dieren. Ook vonden we dat er minder lever-

vervetting optrad in gevaste LXRα knockouts. Het lijkt er daarom op dat LXRα activiteit van belang

is voor de regulatie van metabole fl uxen wanneer het aanbod van voedingsstoff en afneemt, en het

lichaam zijn reserves gaat aanspreken.

De ‘Farnesoid X Receptor’ (FXR) is een transcriptiefactor die geactiveerd wordt door galzouten.

Ook is aangetoond dat FXR knockouts in mindere mate in staat zijn het glucose metabolisme in de

lever op peil te houden wanneer zij gevast worden. De resultaten van de studies die in Hoofdstuk 3 zijn beschreven duiden op een rol voor FXR in de regulatie van het glucose metabolisme in de

dunne darm. De toename in de bloedsuikerspiegel door een orale glucose belasting bleek ver-

minderd te zijn in FXR knockouts. Vervolgexperimenten wezen uit dat de glucose opname in deze

dieren vertraagd was. Dit is te verklaren doordat de passage van glucose door darmcellen in FXR

knockouts gedeeltelijk via een alternatieve route plaatsvindt. Uit deze studies valt te concluderen

dat de inactivatie van FXR leidt tot een verandering van de glucose fl uxen in de darm.

‘Peroxisome Proliferator Activated Receptors’ (PPARs) zijn transcriptiefactoren die worden geac-

tiveerd door vetzuren. PPARα is een belangrijke regulator van het vetmetabolisme in de lever. Zo

is aangetoond dat deze transcriptiefactor een sleutelrol speelt bij de metabole aanpassingen die

plaatsvinden tijdens de overgang van voeden naar vasten. Dit houdt in dat de verbranding van

vetzuren toeneemt, terwijl het glucoseverbruik beperkt wordt. Daarnaast activeert PPARα een aantal

systemen die beschermen tegen schade door eindproducten van vetverbranding. De studies in

Hoofdstuk 4 werpen een nieuw licht op de aanpassingen die optreden als gevolg van een verhoog-

de PPARα activiteit in de lever. Muizen werden behandeld met een geneesmiddel dat PPARα acti-

veert. Als gevolg hiervan was de transcriptie van enzymen die betrokken zijn bij de vetverbranding

in de lever verhoogd. Verrassend genoeg zagen we ook een toename in de transcriptie van enzy-

men benodigd voor vetzuuraanmaak (‘vetzuursynthese’). Voor deze verhoging bleek de aanwezig-

heid van de transcriptiefactor ‘Sterol Regulatory Element Binding Protein 1c’ (SREBP-1c) noodzakelijk.

Deze transcriptionele regulatie vertaalde zich in een toename van de vetzuursynthese-fl ux en een

ophoping van vet in de lever. Ook traden er veranderingen op in de glucose fl uxen. Al met al laten

deze resultaten zien dat de verbranding en aanmaak van vetzuren gelijktijdig kunnen plaatsvinden

in de lever. De toegenomen aanmaak van vetzuren en de ophoping van vet maken waarschijnlijk

onderdeel uit van een mechanisme waarmee de lever zichzelf beschermt tegen schade door vetzu-

ren en hun afbraak producten.

Hoofdstuk 5 en 6 beschrijven de gevolgen van een verhoogde inname van vet via de voeding.

Hiervoor werden twee verschillende soorten vetrijke diëten gebruikt. Het ene dieet bestond uit

rundvet, en had hierdoor een hoog gehalte aan verzadigde vetzuren. Deze verzadigde vetzuren

waren in het andere dieet gedeeltelijk vervangen door visolie, wat rijk is aan meervoudig onverza-

1 5 1Samenvatting

digde vetzuren. De inname van deze vetzuren verlaagt de kans op aderverkalking en hartinfarcten.

De diëten werden gedurende zes weken aan verschillende groepen muizen aangeboden. De ver-

hoogde vetinname leidde in beide gevallen tot een toename van de vetverbranding in vergelijking

tot muizen die een standaard (laag vet) dieet kregen. Dit eff ect was het meest uitgesproken in de

dieren die het met visolie verrijkte dieet aten (Hoofdstuk 5). Toch veroorzaakten beide hoog vet

diëten een toename van de vetmassa, omdat door het hoge vetgehalte de energie-inhoud van de

voeding ook steeg, terwijl het energieverbruik gelijk bleef (Hoofdstuk 5). Dieren die het rundvet

dieet kregen, hadden een verhoogde vetzuursynthese fl ux in hun lever. Dit was mogelijk het gevolg

van een verhoogde transcriptie van vetzuursynthese enzymen (Hoofdstuk 6). Aan de andere kant

was de hoeveelheid vet die de lever verliet gelijk aan dat van muizen die een standaard dieet aten.

Hierdoor was de netto opslag van vet verhoogd in de levers van de met rundvet gevoede muizen.

Gedeeltelijke vervanging van het verzadigd vet door visolie leidde tot een onderdrukking van zo-

wel de transcriptie van vetzuursynthese enzymen als de werkelijke fl ux door deze route. Hierdoor

werd de ophoping van vet voorkomen. Visolie heeft dus gunstige eff ecten op het vetmetabolisme,

die bijdragen aan het voorkomen van leververvetting en hartinfarcten. Tenslotte werden de conse-

quenties van de verhoogde vetinname voor glucose fl uxen bestudeerd. Inname van het rundvetdi-

eet beïnvloedde het glucose metabolisme op negatieve wijze: de gevoeligheid voor insuline nam

af. Hierdoor neemt op de lange termijn het risico op suikerziekte toe. De langdurige overbelasting

met voedingsvet leidde tot een permanente vetverbranding, met als gevolg een daling van de glu-

coseopname en glucoseverbranding. In tegenstelling tot de eerder genoemde verbeteringen in

het vetmetabolisme, resulteerde de visolie verrijking niet tot een verbetering van het glucose me-

tabolisme. Dit verslechterde zelfs ten opzichte van de met rundvet gevoerde dieren (Hoofdstuk 5).

Andere studies laten wel een beschermend eff ect zien van visolie op het ontwikkelen van insuline

resistentie van het glucose metabolisme. Daarom is het van belang de eff ecten van visolie onder

verschillende omstandigheden te bestuderen. Hiervoor is vervolgonderzoek nodig.

Deze studies vergroten het inzicht in de werking van transcriptiefactoren die metabole fl uxen

reguleren. Zo lijkt een verhoogde activiteit van bepaalde factoren noodzakelijk voor een adequate

verwerking van vetzuren. Hierdoor wordt de kans op leverschade verkleind. In vele gevallen ver-

oorzaakt dit echter een ophoping van vet in de lever. Hoewel deze vetstapeling op zichzelf redelijk

onschuldig is, wordt hierdoor het risico op leveraandoeningen vergroot, in het bijzonder wanneer er

gelijktijdig een ontstekingreactie plaatsvindt. Verder laten de studies zien dat een blokkade van de

werking van transcriptiefactoren leidt tot een reorganisatie van metabole fl uxen. Deze reorganisatie

is vaak veel ingrijpender dan wat verwacht kan worden op basis van de directe regulatie door target

genen van een bepaalde transcriptiefactor. Bovendien kan het metabolisme door het samenspel

van enzymen op verschillende manieren beïnvloed worden onder wisselende metabole omstan-

digheden. Het is daarom van groot belang om de eff ecten van transcriptiefactoren binnen de juiste

context te bestuderen. Tenslotte kan een verandering van de ene metabole fl ux gevolgen hebben

voor een andere fl ux. Het toepassen van ‘fl uxomics’ in intacte organismen is daarom nodig om de

eff ecten van medicijnen die aangrijpen op transcriptiefactoren volledig in beeld te krijgen. Wan-

neer de globale metabole gevolgen in kaart worden gebracht zullen uiteindelijk betere medicijnen

ontwikkeld kunnen worden.

1 5 3Dankwoord

Dankwoord

‘Promoveren is als Biochemie’

Er zijn momenten dat je als AIO het gevoel hebt dat je met een ‘solo-actie’ bezig bent. Toch is on-

derzoek doen wel degelijk een teamsport. Waar biochemische reacties mogelijk gemaakt worden

door enzymen, waren het tijdens mijn promotieonderzoek de door mij bestempelde ‘helden’ en

‘koningen’ die het proces katalyseerden.

Dirk-Jan, ik kreeg van jou veel vrijheid, wat ik heerlijk, maar soms ook lastig vond. Uiteindelijk heeft

dit volgens mij goed uitgepakt, en ertoe geleid dat er wel degelijk wat scherpe kantjes zijn ontstaan.

Onze besprekingen leken zo nu en dan plaats te vinden onder het motto ‘wie het hardst schreeuwt

heeft gelijk’, maar dat is onvermijdelijk wanneer je twee parkieten bij elkaar zet. Ik heb veel van je

geleerd, en realiseer me dat de biochemie steeds meer tot mijn verbeelding gaat spreken.

Folkert, hoewel ik als voedingsmiepje nauwelijks ervaring had met moleculaire biologie en farma-

cologie, was het jouw fysiologische kijk die me aansprak en inspireerde. Bovendien zorgden jouw

rust, nuchterheid en heldere visie (‘SHIT! -Diabetes gaat niet lukken-F’) ervoor dat ik gemotiveerd

bleef. Ik denk met veel plezier terug aan de operatieuren tussen-de-vier-muren van ADL08, en de

door jou geïmproviseerde anglicismen van Nederlandse uitdrukkingen tijdens werkbesprekingen.

Maar bovenal ben ik je ontzettend dankbaar voor de ruimte en het vertrouwen die ik kreeg om mijn

eigen weg te banen.

The Manuscript Committee: dear profs. Frayn, Romijn and Groen, thank you for your willingness to

judge my work, and for your rapid approval of this dissertation.

Bert, bedankt voor de frisse wind door mijn manuscripten. Het is top dat je er bent!

theo, zullen we de ketenverlenging nog één keer herberekenen ;)? Jij bent degene die in praktische

zin het meest geconfronteerd werd met de controversiële resultaten (en dat vond jij best zolang het

maar niet jouw experiment was). Dus puzzelde je mee om nieuwe experimenten te verzinnen en

om de bestaande isotoop-methoden nóg beter te maken. Bovendien was het samen-proeven-doen

altijd erg gezellig. Aldo, we hebben regelmatig intensief samengewerkt en dat is me erg goed beval-

len. In het begin hobbelde ik braaf achter je aan, en leerde zo het klappen van de zweep kennen.

Later, toen jij terugkeerde uit Dallas dook je in mijn fenofi braat-avontuur en kreeg je hier een belang-

rijke rol in. Gelukkig zijn we voorlopig nog lang niet uitgeklust met onze gemeenschappelijke hobby

LXR. Rick, een dag samen in de OK staat gelijk aan het verenigen van nuttige en aangename zaken.

Dank voor de gezellige uurtjes achter de microscoop. THEO, mede dankzij jouw spectometrie-oog

is de ketenverlenging-MIDA een succes geworden. Op naar de volgende uitdaging! Trijnie, hoewel

we ‘pas’ sinds een jaar of anderhalf ons LPLeed delen, heb ik het gevoel dat we al veeeeeel langer

samenwerken. Dank voor je fi jne hulp en vrolijke noot.

1 5 4 Dankwoord

Daarnaast hadden ook de volgende helden een rol in mijn promotie-avontuur:

De medewerkers van het CDP, in het bijzonder Diana, Harm, Ralph, Ar, Lucas en Natasha: dankzij

jullie werden de dieren goed verzorgd, en leken de dagen op ADL08 een stuk minder lang. Ingrid

(vetzuurkoningin) en Claude, bedankt voor jullie hulp bij het GC-en. Margriet en Ko, dank voor jul-

lie input in het visolie-glucose stuk. Sjoerd, (een echte ingenieur, de SPSS koning), met plezier denk

ik terug aan onze telefonische brainstormsessies. Gertjan van Dijk en Bart Staels, bedankt voor de

inspirerende samenwerking.

De rest van het ‘Kindergeneeskunde/Metabole Ziekten/MDL-lab’: de fi jne collega’s die voor de goe-

de sfeer zorgen, waardoor ik me snel thuis voelde in Groningen, en die me niet vergaten nadat ik per

ongeluk een gat in gedoken was: Sabina (vetzuurprinses, tevens fashion police, bedankt dat je me

naar huis stuurde toen mijn outfi t echt niet kon), Maxi (nice earrings), Hilde (alleen jij mag me Henk

noemen), AnkeR (binnenkort weer ‘ns een paal opzoeken?), Annelies (ik kijk tegenwoordig uit voor

overstekende stoepranden), Gemma (thanks for keeping an eye on me in SF), Uwe (too bad I didn’t

make it into your group), Jan-Fraerk (met stadsfi ets de Cauberg bedwingen = respect!), Marije, Jurre,

Torsten, Harmen, Juul, Nicolette, Wytse, Frans C, Hester, Renze, Vincent (bedankt voor het scherp

houden), Frans S, Janine, Yan, Mark, Thomas, Henkjan (I just need an answer), Hilde & Gea (onvermij-

delijk, onmisbaar en onverbiddelijk), Albert (ik wou soms dat ik jou was), pim, Fjodor, Klaas, Elles, Ti-

neke, PIM, Ingrid, Annet, Jenny, Conny, Marianne, Hermi, Janneke, Stieny, Aicha, Mariette, Willemien,

Marjan, Agnes, Barbara, Rebecca, Dolf, Frank P, Janny, Feike, Karin, Robert (jij bent echt slim), Thierry,

Alberto, Anja, Renate, Terry, Edmond, Frank B, Marion, Han, Roel, Janneke, Marianne, Ewa, Mariska,

Manon, Jannes, Axel, Tjasso, Elise, Fiona, Janette, Krzysztof, Laura, Antonella, Rebekka, Sandra, Golnar,

Atta, Rohina, Anouk, Klaas-Nico en Han.

Mijn kamergenoten: bedankt voor het tolereren van mijn aan- en afwezigheid, voor de opbeurende

woorden en de gezellige kletspraat. Esther (miepje-in-crime, binnenkort 2x dr. ir), Niels (allerliefste

über-held, koffi e?), Hans en Martijn (de hanen), Henk (die helaas moest ondervinden hoe lastig het

is om het insuline signaal te signaleren), Anniek, Miriam, Wytske, Arne, Jaana en Margot.

Last but not least mijn ‘mede-kipjes’, het gezelschap dat elkaar zonder te spreken verstaat:

Leo en Titia: het was een dolle boel, we schateren vrolijk door in Nijmegen! Jelske: anderhalf jaar

voelde als één dag, wat is de bestemming van onze volgende stedentrip?

Lieve paranifmen, Anniek: ondanks dat www.femalesforfree.com het niet gered heeft, was ons Ha-

ren-avontuur achteraf gezien ontzettend leuk en bovendien erg stoer. Dat kwam in de eerste plaats

door ons geweldige team (met Pieter en Nanda ‘komt goed!’). Dat we het overleefd hebben is voor

een deel te danken aan jouw persoonlijkheid, ik bewonder je incasseringsvermogen en rust. De

resultaten van ons mega-project staan dan wel niet in dit proefschrift, maar ik ben ervan overtuigd

dat ze binnen de kortste keren wereldkundig zullen zijn! Marijke: hospita, contragewicht, gedach-

tendievegge, hulplijn, lotgenoot, voedselbank, luisterend oor en geweten. Vanaf het moment dat jij

en Jelle zo gastvrij waren onderdak te bieden aan een licht-gehandicapte Maaike met bijbehorend

elleboog-oefenapparaat, loopt jouw aanwezigheid als een rode draad door mijn AIO-tijd. Dat de

wetenschappelijke vruchten van het M&M project op zich laten wachten, is simpelweg te verklaren

doordat zij gek zijn. Zo niet dan toch, enne, alles komt goed!

1 5 5Dankwoord

Alle helden-op-afstand: wat fi jn dat jullie zo betrokken waren en interesse toonden!

Mirjam, Leanne, Bart, Lara, Jantine, Bart, Jolieke, Philip, Jasper en patatkippen: vriendschap is dimen-

sieloos! Vesta: Jils, Jet, Veer, Janni, Thinus, M’riek, Daan, An en Sint; na al die jaren is ons vuurtje verre

van gedoofd!

Dineke en Johan, ik voel me onvoorwaardelijk omarmd. Bedankt voor jullie support, op zoveel ver-

schillende manieren en vlakken! Daniel: ik vind het fi jn dat jij wel om mijn grappen lacht. Heb ik met

het afronden van dit proefschrift nu bevestigd dat ik niet slim ben, maar gewoon heel hard kan

werken? Joris, Freek en Wouter, mijn ‘kleine’ broertjes: ik ben trots op jullie! (jullie ook een beetje op

mij?). Lotte, Chantal en Michelle: het is toch nog goed gekomen met die broers van mij ;). Houd ze

in de gaten! Pap en mam, jullie hadden de heldenstatus allang bereikt. Doordat jullie ons alle kansen

en ruimte hebben geboden, zijn we geworden wie we zijn, en dat is zo slecht nog niet! De weken

in Wyler hebben me ontzettend geholpen om de chaos in mijn hoofd de baas te worden, en mijn

gedachten dusdanig te ordenen zodat ze ook voor anderen begrijpelijk werden. Dank voor jullie

liefde, steun en fl exibiliteit.

Michiel, jij bent zonder twijfel de grootste held! Het is fi jn om met iemand samen te leven die min-

stens zo kritisch is als ikzelf. Jouw rotsvaste vertrouwen in mijn kunnen motiveert me om telkens

een tandje bij te schakelen. Daarnaast helpt jouw down-to-earth mentaliteit me om dingen te nu-

anceren wanneer ik weer eens dreig door te draven. In het afgelopen jaar hebben we behoorlijk

wat teamprestaties geleverd, en wanneer ik me realiseer wat we samen hebben bereikt voel ik me

ontzettend blij en trots. Met jou is de wereld mooier!

1 5 7Biografi e/Biography

Biografi e

Maaike Hélène Oosterveer werd op 10 juli 1980 geboren te Nijmegen. Hier behaalde zij in 1998

haar Atheneumdiploma aan het Kandinsky College. Hetzelfde jaar startte zij de studie Voeding en

Gezondheid aan Wageningen Universiteit. Haar eerste afstudeeronderzoek (2002) vond plaats bij

de vakgroep Fysiologie van Mens en Dier (Wageningen Universiteit; supervisie: dr. B.J. van de Heij-

ning en prof. dr. D. van der Heide). In 2003 verbleef zij in het kader van een onderzoeksstage aan

het Rowett Research Institute (supervisie: dr. H.S. Andersen, dr. C. Fosset en prof. dr. H.J. McArdle)

gedurende vier maanden in Aberdeen, UK. Bij terugkeer in Nederland besloot ze haar studie te ver-

lengen met een tweede afstudeeronderzoek aan Wageningen Universiteit bij de vakgroep Humane

Voeding en Epidemiologie (supervisie: dr. H.M. van den Bosch, dr. ir. G.J. Hooiveld en prof. dr. M.

Müller). Eind 2003 behaalde zij haar ingenieursgraad. In maart 2004 startte ze haar promotietraject

bij het Researchlaboratorium Kindergeneeskunde van het Universitair Medisch Centrum Groningen.

Haar promotieonderzoek vond plaats binnen het kader van een door het Diabetes Fonds Nederland

gesubsidieerd project getiteld ‘Molecular basis of fi sh-oil prevention of type 2 diabetes’ onder super-

visie van dr. D-J. Reijngoud en prof. dr. F. Kuipers. De resultaten van dit onderzoek zijn beschreven in

dit proefschrift.

Sinds november 2008 is ze bij het Researchlaboratorium Kindergeneeskunde aangesteld als post-

doctorale onderzoeker. Hier is ze werkzaam binnen een door het Top Instituut Pharma gesubsidi-

eerd onderzoeksproject getiteld ‘Nuclear receptors as targets for anti-atherosclerotic therapies’.

Biography

Maaike Hélène Oosterveer was born in Nijmegen on the 10th of July, 1980. She graduated from high

school in 1998 and started the education programme Nutrition and Health at Wageningen Univer-

sity. She performed her fi rst research project at the Department of Human and Animal Physiology

(Wageningen University, supervisors: dr. B.J. van de Heijning and prof. dr. D. van der Heide). In 2003,

she spent four months in Aberdeen (UK) for a research internship at the Rowett Research Institute

(supervisors: dr. H.S. Andersen, dr. C. Fosset and prof. dr. H.J. McArdle). After her return in the Net-

herlands, she started a second research project at Wageningen University (Department of Human

Nutrition and Epidemiology, supervisors: dr. H.M. van den Bosch, dr. G.J. Hooiveld and prof. dr. M.

Müller). At the end of 2003, she received her Masters degree. In March 2004, she started her PhD at

the Laboratory of Pediatrics of the University Medical Center Groningen, under supervision of dr. D-J.

Reijngoud and prof. dr. F. Kuipers. Her PhD project entitled ‘Molecular basis of fi sh-oil prevention of

type 2 diabetes’, was funded by the Dutch Diabetes Research Foundation. The results of this research

are summarized in this dissertation.

As from November 2008, she is appointed as a post-doctoral researcher in the Laboratory of Pedi-

atrics on a project entitled ‘Nuclear receptors as targets for anti-atherosclerotic therapies’, which is

funded by the Top Institute Pharma.

1 5 8 Publications

List of PublicationsLxrα defi ciency hampers the hepatic adaptive response to fasting in mice.Oosterveer MH, van Dijk TH, Grefhorst A, Bloks VW, Havinga R, Kuipers F, Reijngoud D-J.

J Biol Chem. 2008 Sep 12;283(37):25437-45.

An increased fl ux through the glucose 6-phosphate pool in enterocytes delays glucoseabsorption in Fxr -/- mice.van Dijk TH*, Grefhorst A*, Oosterveer MH, Bloks VW, Staels B, Reijngoud D-J, Kuipers F.

J Biol Chem. 2009 Apr 17;284(16):10315-23.

Fenofi brate simultaneously induces hepatic fatty acid oxidation, synthesis and elongation in mice.Oosterveer MH, Grefhorst A, van Dijk TH, Havinga R, Staels B, Kuipers F, Groen AK, Reijngoud D-J.

Conditionally accepted for publication

Fish oil potentiates high-fat diet-induced peripheral insulin resistance in mice.Oosterveer MH, Schreurs M, van Dijk TH,Wolters H, Havinga R, van den Berg SAA, Willems van Dijk K,

van der Zon GCM, Ouwens DM, Groen AK, Kuipers, F, Reijngoud D-J.

Submitted

High fat feeding induces hepatic fatty acid elongation in mice.Oosterveer MH, van Dijk TH, Tietge UJF, Boer T, Havinga R, Stellaard F, Groen AK, Kuipers F, Reijngoud

D-J.

PLoS ONE. 2009 Jun 26;4(6):e6066.

Peroxisome proliferator-activated receptor alpha improves pancreatic adaptation to insulin resistance in obese mice and reduces lipotoxicity in human islets.Lalloyer F, Vandewalle B, Percevault F, Torpier G, Kerr-Conte J, Oosterveer MH, Paumelle R, Fruchart JC,

Kuipers F, Pattou F, Fiévet C, Staels B.

Diabetes. 2006 Jun;55(6):1605-13.

Pharmacological inhibition of the acetyl-CoA carboxylase system by CP-640186 improves peripheral insulin sensitivity in mice.Schreurs M, Oosterveer MH, van Dijk TH, Gerding A, Havinga R, Reijngoud D-J, Kuipers, F.

Submitted

Metabolic responses to long-term pharmacological inhibition of CB1-receptor activity in mice in relation to dietary fat composition.Koolman AH, Bloks VW, Oosterveer MH, Jonas I, Kuipers F, Sauer PJJ, van Dijk G-J.

Conditionally accepted for publication

Resistance to diet-induced obesity in CB1-receptor defi cient mice is not related to impaired lipogenesis or lipolysis in adipose tissue.Oosterveer MH*, Koolman AH*, de Boer PT, Bos T, van Dijk TH, Havinga R, Kuipers F, van Dijk G-J.

In preparation

1 5 9Publications

Resistance of CB1-receptor defi cient mice to diet-induced hepatic steatosis can not be attributed to suppressed lipogenesis in liver.Koolman AH*, Oosterveer MH*, Bos T, Havinga R, van Dijk TH, Sauer PJJ, Kuipers F, van Dijk G-J.

In preparation

Postnatal regulation of weight gain by endocannabinoid signalling in mice.Koolman AH, Gruben N, Oosterveer MH, Sauer PJJ, Kuipers F, van Dijk G-J.

Submitted

Bile salt sequestration induces de novo lipogenesis by impaired hepatic bile salt signalling.Herrema HJ, Meissner M, van Dijk TH, Brufau G, Boverhof R, Oosterveer MH, Reijngoud D-J, Müller M,

Stellaard F, Groen AK, Kuipers F.

In revision

* these authors contributed equally

Date post:	19-Feb-2018
Category:	Documents
Upload:	nguyenkhuong
View:	217 times
Download:	2 times

University of Groningen Control of metabolic flux by ... · PDF fileControl of Metabolic Flux...

Documents