uva-dare (digital academic repository) the metabolic ... · table of contents chapter 1...
TRANSCRIPT
UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
The metabolic response to fasting in humans: physiological studies
Soeters, M.R.
Link to publication
Citation for published version (APA):Soeters, M. R. (2008). The metabolic response to fasting in humans: physiological studies.
General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.
Download date: 02 Jul 2019
one cannot comprehend the pathophysiology of insulin resistance, when the physiological sense of the metabolic adaptations to fasting is not understood
maarten rené soeters was born during a heat wave on august 2nd
1975 in boston, massachusetts (u.s.a). he graduated from high
school in 1993 and started medical training at the university of
amsterdam right away. as a student he was active in rowing and
cycling. he finished medical training on december 21st 2001. despite
surgical temptations, he became a resident internal medicine in the
kennemer gasthuis in haarlem. after 10 months he started internal
medicine training at the academic medical center in amsterdam.
in 2004 he joined the metabolic group of prof. dr. h.p. sauerwein,
by now dr. m.j.serlie. this resulted in a phd program for which he
postponed his internal medicine training between october 2005
and october 2007. on april 1st 2008 he started his endocrinology
and metabolism fellow-ship under supervision of prof. dr. e. fliers.
he is married to hanny hamringa and they have four children.
the metabolic response to fasting in humansphysiological studies
maarten r. soeters
the metabolic response to fasting in hum
ans m
aarten r. soeters
proefschrift maarten soeters v2.0_PS_010808.indd 1 3-8-2008 22:21:00
stellingen the metabolic response to fasting in humans physiological studies
1. tijdens vasten worden vrouwen beschermd tegen insuline resistentie
geïnduceerd door vrije vetzuren (dit proefschrift) 2. proteïne kinase b/akt phosphorylering (serine473) is van belang voor
vasten geïnduceerde insuline resistentie (dit proefschrift) 3. de aanpassing van de vetzuuroxidatie tijdens vasten suggereert dat er
sprake is van een veranderd set-point (dit proefschrift) 4. beta-hydroxyboterzuur wordt niet obligaat geoxideerd tijdens vasten: er
is tevens sprake van hydroxybutyrylcarnitine-synthese, een tot nu toe onbekend metabool pad (dit proefschrift)
5. gelijke remming van ketogenese door insuline in slanke en obese
mannen pleit voor differentiële insuline gevoeligheid van de intermediaire stofwisseling (dit proefschrift)
6. intermitterend vasten heeft gunstige effecten op glucose, vet en eiwit
metabolisme, maar niet op gewicht (dit proefschrift). 7. plasma glucose waarden kleiner dan 3.0 mmol/L zijn niet per sé
afwijkend (dit proefschrift) 8. er is onvoldoende evidence dat het harde smalle zadel van een racefiets
leidt tot libidoverlies of infertiliteit van de wielrenner (eigen waarneming) 9. de body mass index (kilogram/meter2) is een magere maat omdat de
formule suggereert dat gebrek aan lichaamslengte een oorzaak kan zijn van obesitas
10. les admirations ne sont vraies comme les amitiés que lorsqu'elles laissent
libres les opinions (bernard halda, ± 1930) 11. stellige uitspraken zijn vaak gewoonweg onwaar (behalve deze)
maarten r. soeters, 11 september 2008
the metabolic response to fasting in humans
physiological studies
proefschrift Soeters.indb 1 4-8-2008 9:13:15
The Metabolic Response to Fasting in Humans
Physiological Studies
Dissertation, University of Amsterdam
Copyright© 2008, Maarten R. Soeters, Amsterdam the Netherlands
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.
Author: Maarten R. Soeters
Cover: tafel met gat, 2008 by Peter-Jan Soeters and Merijn B. Soeters
Lay-out: Chris D. Bor
Print: Buijten en Schipperheijn, Amsterdam
ISBN 978-90-813298-1-1
This dissertation was funded by Stichting Amstol, Ferring BV Hoofddorp, Genzyme,
Goodlife, GlaxoSmithKline, Novo Nordisk, Nutricia Advanced Medical Nutrition, Sanofi
Aventis and the University of Amsterdam
proefschrift Soeters.indb 2 4-8-2008 9:13:15
the metabolic response to fasting in humans
physiological studies
Academisch Proefschrift
ter verkrijging van de graad van doctor
aan de Universiteit van Amsterdam
op gezag van de Rector Magnificus
prof. dr. D.C. van den Boom
ten overstaan van een door het college voor promoties ingestelde
commissie, in het openbaar te verdedigen in de Aula der Universiteit
op donderdag 11 september 2008, te 14.00 uur
door
Maarten René Soetersgeboren te Boston, Verenigde Staten van Amerika
proefschrift Soeters.indb 3 4-8-2008 9:13:15
Promotiecommissie
Promotor: Prof. dr. H.P.Sauerwein
Co-promotores: Dr. M.J. Serlie
Dr. ir. M.T.Ackermans
overige leden: Prof. dr. E. Blaak
Prof. dr. F. Kuiper
Prof. dr. M.M. Levi
Prof. dr. J.A. Romijn
Prof. dr. W.M. Wiersinga, emeritus
Prof. dr. F.A. Wijburg
Faculteit der Geneeskunde
proefschrift Soeters.indb 4 4-8-2008 9:13:15
Voor Hanny
proefschrift Soeters.indb 5 4-8-2008 9:13:15
proefschrift Soeters.indb 6 4-8-2008 9:13:15
Table of Contents
Chapter 1 Introduction. 9
Chapter 2 Gender related differences in the metabolic response to fasting. 27
Chapter 3 Muscle adaptation to short-term fasting in healthy lean humans. 45
Chapter 4 Decreased suppression of muscle fatty acid oxidation by
hyperinsulinemia during fasting.
57
Chapter 5 Muscle D-3-hydroxybutyrylcarnitine: an alternative pathway in
ketone body metabolism.
73
Chapter 6 Equal insulin sensitivity on inhibition of ketogenesis during short
term-fasting in lean and obese humans.
89
Chapter 7 Intermittent fasting differentially affects glucose, lipid and protein
metabolism.
105
Chapter 8 Hypoinsulinemic hypoglycemia during fasting in adults:
a Gaussian Tale.
123
Chapter 9 The metabolic response to fasting in humans: a Perspective. 141
Chapter 10 English summary. 157
Nederlandse Samenvatting. 163
Dankwoord. 167
Nederlandse Biografie. 172
proefschrift Soeters.indb 7 4-8-2008 9:13:15
proefschrift Soeters.indb 8 4-8-2008 9:13:15
GENERAL INTRODUCTION1
proefschrift Soeters.indb 9 4-8-2008 9:13:15
proefschrift Soeters.indb 10 4-8-2008 9:13:15
General introduction
11
Chapter 1
1 INTRODUCTION
General considerationsFasting has been part of human’s nature since the very beginning of mankind: the
adaptations that occur in the fasting state serve to protect the organism from substrate
depletion. During fasting the plasma glucose concentration needs to be kept in narrow
limits to fuel the central nervous system. In addition, adaptations to preserve protein mass
will increase our chances of survival. The human organism has extensive fuel reserves,
mainly represented by adipose tissue and muscle, as shown in table 1 (1).
Table 1 Fuel reserves in a typical 70-kg man
Available energy in kcal (kJ)
Organ Glucose/glycogen Triacylglycerols Mobilizable protein
Blood 60 (250) 45 (200) 0 (0)
Liver 4 (1700) 450 (2000) 400 (1700)
Brain 8 (30) 0 (0) 0 (0)
Muscle 1200 (5000) 450 (2000) 2400 (100000)
Adipose tissue 80 (330) 135000 (560000) 40 (170)
Adapted from: G.F. Cahill Jr, Clin. Endocrinol. Metab. 5 (1976):398. With permission from Elsevier.
Fasting can be defined as a lower energy intake than needed to maintain body weight,
however in this thesis, the term fasting is reserved for the complete withdrawal of food
except for water.
Tight definitions for different durations of fasting are lacking, however after critical
appraisal of the literature some statements can be made regarding this topic.
Most known is overnight fasting (defined as 14 h fasting or post-absorptive state) which
is being used extensively to prepare patients for interventions or diagnostic procedures.
Surgery is most often scheduled after an overnight fast to prevent pulmonary aspiration
although the evidence for this practice may be challenged (2). Screening for diabetes
mellitus with a fasting plasma glucose level or oral glucose tolerance test typically occurs
after an overnight fast (3), in order to compare the well defined response to 75 gram
of glucose without fluctuations in plasma insulin and glucose caused by intestinal food
absorption.
When fasting is continued after an overnight fast, this is named short-term fasting.
Most studies that have examined the physiological adaptations up to 85 h of fasting, define
such fasting periods as short-term (4-8). Although fasting for 85 hours unequivocally is
proefschrift Soeters.indb 11 4-8-2008 9:13:15
12
a very long time, it is actually quite understandable why this is called short-term fasting.
Short-term fasting encomprises the first adaptive changes that occur in the adaptation in
energy metabolism (hypoinsulinemia, decreased glucose oxidation, increased fatty acid
oxidation (FAO), increased lipolysis, ketogenesis, decreased insulin mediated glucose
uptake, glycogenolysis (and subsequent depletion of glycogen stores), gluconeogenesis
and proteolysis (5;7-17). These adaptations are active in the post-absorptive state but
become maximal after approximately 3 days of fasting (5)
The total energy reserves in a typical 70 kg man represent 161000 kcal, as shown in
table 1 (1). The resting energy expenditure (REE) then may be approximately 1600 to
1700 kcal (18). This means that energy stores will suffice to meet caloric needs for 95 to
101 days (19). However this is a simplified example because it does not take into account
activity, changes in REE and critical protein mass for survival (20).
Intermittent fasting is another way of fasting that is most easily exemplified by the way
Muslims fast during the Ramadan (21). Intermittent fasting is characterized by refraining
form food intake in certain intervals and was studied because of the thrifty gene concept
(22). Although the thrifty gene concept has been debated (23;24), animal and human
studies have shown effects of IF on energy metabolism (25). Previous human studies
have not been numerous and did not yield equivocal data (22;26;27). However, the
increase of insulin sensitivity after two weeks of IF shown by Halberg et al is of interest
since it may provide in a simple tool to improve insulin sensitivity (22).
However it is unknown whether eucaloric IF affects basal endogenous glucose
production (EGP), hepatic insulin sensitivity or protein breakdown in healthy lean
volunteers.
Metabolic adaptation to short-term fastingFatty acid metabolism during short-term fasting
During short-term fasting, the orchestrated interplay of low insulin levels and increased
plasma concentrations of catecholamines and growth hormone (GH) will increase
lipolysis and thus plasma free fatty acid (FFA) concentrations (5-9;14;28). Lipolysis is
activated by protein kinase A (PKA) and inhibited by insulin (29). PKA phosphorylates
hormone sensitive lipase (HSL) and perilipin A, leading to translocation of HSL to lipid
droplets (30-32). Adipose triglyceride lipase (ATGL) is another lipase involved in hydrolysis
of triglycerides (33).
During short-term fasting, fatty acid oxidation (FAO) increases with a concomitant
decrease of glucose oxidation (CHO) (6;34). Fasting increases the intramyocellular
AMP/ATP ratio that activates AMP-activated protein kinase (AMPK) by AMP binding
proefschrift Soeters.indb 12 4-8-2008 9:13:15
General introduction
13
Chapter 1
and phosphorylation (35). Phosphorylated AMPK then phosphorylates and inhibits ACC
activity, thereby inhibiting malonyl-CoA synthesis. This results in decreased inhibition of
carnitine-palmitoyltransferase 1 (CPT-1) activity, thereby increasing mitochondrial import
of fatty acids and FAO in muscle. The increase in FAO will yield acetyl-CoA which inhibits
glycolysis and subsequent glucose oxidation via the pyruvate dehydrogenase complex
(PDHc). In a way, this has been described by Randle in his renowned paper on the glucose
fatty-acid cycle in 1963 (13). Randle’s paper was concluded as follows:
“Evidence is presented that a higher rate of release of fatty acids and ketone bodies for oxidation
is responsible for abnormalities of carbohydrate metabolism in muscle in diabetes, starvation, and
carbohydrate deprivation, and in animals treated with, or exhibiting hypersecretion of, growth
hormone or corticosteroids. We suggest that there is a distinct biochemical syndrome, common to
these disorders, and due to breakdown of glycerides in adipose tissue and muscle, the symptoms of
which are a high concentration of plasma non-esterified fatty acids, impaired sensitivity to insulin,
impaired pyruvate tolerance, emphasis in muscle on metabolism of glucose to glycogen rather
than to pyruvate, and, frequently, impaired glucose tolerance. We propose that the interactions
between glucose and fatty-acid metabolism in muscle and adipose tissue take the form of a cycle,
the glucose fatty-acid cycle, which is fundamental to the control of blood glucose and fatty-acid
concentrations and insulin sensitivity.”(13)
This assumption seems more or less correct in view of changes in substrate oxidation
since glucose oxidation is inhibited via PDHc in both starvation and obesity induced insulin
resistance/diabetes mellitus type 2 (34;36). In contrast, the Randle cycle was revisited by
Boden and Shulman by argumenting that in peripheral insulin resistance, the decreased
transmembrane glucose transport is rate limiting and not accumulation of glucose-6-
phosphate as he suggested in his original paper (37).
During fasting, myocellular lipid supply exceeds FAO(16;38). This means that the
excess muscle lipid must be stored in some way. It was shown by an elegant tracer study
that the higher muscle lipid uptake (compared to FAO) is accompanied by increased
reesterification of FFA to triglycerides (38). This is supported by a proton magnetic
resonance spectroscopy (MRS) study which showed accumulation of IMCL in the vastus
lateralis muscle of healthy men during short-term fasting (16).
Long-chain fatty acid transport across the plasma membrane involves a protein-
mediated process by the fatty acid transporters (FAT)/CD36 and fatty acid binding
protein (FABPpm) that are regulated by stimuli like insulin and AMPK (39). After cellular
uptake of plasma FFA, fatty acids are activated to (fatty) acyl-CoA. Long-chain acyl-CoAs
proefschrift Soeters.indb 13 4-8-2008 9:13:16
14
can only transverse the mitochondrial membrane as acylcarnitines (ACs) to be oxidized
(40). The coupling of an activated long-chain fatty acid to carnitine (3-hydroxy-4-N,N,N-
trimethylaminobutyric acid) is catalyzed by CPT1 on the outer mitochondrial leaflet.
Inside the mitochondrion, the fatty acid moiety is activated to acyl-CoA again by CPT2.
The released carnitine is transported to the cytosol again in exchange for a new incoming
AC (by the mitochondrial membrane protein carnitine-acylcarnitine-translocase (CACT)).
Measurement of the ACs in plasma is the golden standard for the diagnosis of FAO
disorders at the metabolite level ((41;42). However, little is known about intramyocellular
AC profiles and their role in local fatty acid and glucose metabolism. Long-chain ACs are
most of interest since metabolites of long chain fatty acids are suggested to interfere
with insulin sensitivity (43-46).
The accumulation of long-chain ACs in plasma may be accompanied by a concomitant
accumulation of muscle ACs levels that could be related to changes in peripheral insulin
sensitivity during fasting (47-49).
Ketone body metabolism during short-term fasting
Plasma ketone body (KB) levels are an important source of energy and their levels and
turnover increase during fasting (1;50). It is somewhat surprising that, until 1967, KBs
have been regarded as “metabolic garbage” with no beneficial physiological role (51).
However, the central nervous system requires approximately 140 g of glucose per day
(equivalent to almost 600 kcal) in which must be foreseen during fasting. Here, KBs are
an excellent respiratory fuel: 100 g of D-3-hydroxybutyrate (one of the KBs) yields 10.5
kg of ATP whereas 100 g of glucose only yields 8.7 kg of ATP.
Ketogenesis (production of the KBs D-3-hydroxybutyrate, acetoacetate and acetone)
occurs in the liver. Here fatty acids undergo beta-oxidation to form acetyl CoA which
enters ketogenesis as depicted in figure 1 (19). Mitochondrial 3-hydroxy-3-methylglutaryl-
CoA synthase (mHMG-CoA synthase) is the major enzyme involved in ketogenesis and is
inhibited by insulin (52). It has been suggested that insulin resistance on ketogenesis, i.e.
less insulin-mediated suppression of ketogenesis, is present in type 2 diabetes mellitus
patients (53).
Whether the KB production rate is different under equal plasma insulin levels in lean
and obese ketotic men is currently unknown.
During ketolysis (KB oxidation) in target organs, the ketogenesis pathway is reversed
as depicted in figure 2. 3-ketoacyl-CoA transferase is not found in the liver; hence hepatic
ketolysis does not exist.
proefschrift Soeters.indb 14 4-8-2008 9:13:16
General introduction
15
Chapter 1It has been suggested by at least two separate studies that 3-hydroxybutyrylcarnitine
(3HB-carnitine) is the product of the coupling of D-3HB and carnitine (44;54). Although it
has been described that D-3HB can be activated to D-3HB-CoA in rat liver and hepatoma
cells, this has not been established in humans (55-57). Furthermore, it is unknown
whether and how D-3HB-CoA can be coupled to carnitine resulting in D-3HB-carnitine.
This is in contrast to its generally known stereo isomer L-hydroxybutyryl-CoA (L-3HB-
CoA), formed via beta-oxidation of butyryl-CoA (58). Although it was proposed that the
total amount of tissue hydroxybutyrylcarnitine was derived from D-3HB and carnitine, the
quantative contributions of the L and D stereo isomers were not accounted for in these
studies (44;54).
Therefore it is unraveled which stereo isomers are present in vivo during short-term
fasting in skeletal muscle and whether coupling of activated D-3HB to carnitine occurs.
Glucose metabolism during short-term fasting
Endogenous glucose production
It is generally known that plasma glucose levels decrease during fasting (4;8;9). This is
explained by the slowly decreasing EGP (4) which is the greatest denominator of the
fasting plasma glucose level (59). In resting circumstances glycogen stores will be reduced
to a minimum after approximately 40h of fasting after which endogenous glucose
production (EGP) primarily relies on GNG (12;60). Furthermore 80-90% of EGP is covered
Figure 1, Ketogenesis
Figure 2, Ketolysis
proefschrift Soeters.indb 15 4-8-2008 9:13:16
16
by hepatic glucose production whereas 10-20% is provided by renal glucose production
(mainly from lactate) (61). It is interesting that women display lower plasma glucose
levels during short-term fasting compared to matched men (8;9). EGP has not shown
to be different between men and women after short-term fasting (4;8;9;62), but one
study demonstrated a steeper decline in time of EGP in women compared to men (4).
In general, healthy humans do not become clinically hypoglycemic (i.e. neuroglycopenic)
because of the contraregulatory response (63;64) and subsequent changes in glucose
and fat oxidation as discussed above. When fasting-induced hypoglycaemia occurs, most
patients undergo a supervised fasting period to exclude hyperinsulinemic hypoglycaemia.
Some subjects develop a fasting-induced hypoglycaemia without neuroglycopenic
symptoms in the presence of low insulin levels. This is a challenging clinical problem.
It is unknown whether patients with fasting-induced plasma glucose levels that fall
well below the threshold value for hypoglycemia (i.e. 2.8 mmol/liter)(63) in the absence
of hyperinsulinemia and signs of neurohypoglycemia have a metabolic disorder (FAO
disorder, low EGP etc.) which could explain the lower tolerability to fasting
The hormonal contraregulatory response is characterized by stimulating effects of
glucagon, growth hormone, cortisol and catecholamines on the key gluconeogenic
enzymes that have been reviewed earlier (65).
The influencing role of FFA on EGP in fasting humans remains enigmatic (66;67).
Despite the reciprocal changes in GGL and GNG during fasting, manipulation of plasma
FFA had little or no effect on the EGP (68). Collected data in fasting humans (up to 40
h) suggest that plasma FFA increase, thereby leaving absolute GNG unchanged (66). In
contrast, Féry et al demonstrated that FFA lowering by acipimox after 104 h of fasting
increased GNG (69) which may reflect the need for oxidative fuel during FFA depletion.
In this respect it is of interest that women have higher plasma FFA compared to men
during fasting (4;8;9).
The explanation for lower plasma glucose levels with higher plasma FFA in fasting
women is still lacking.
Peripheral insulin stimulated glucose uptake during fasting
Under insulin stimulated circumstances, glucose disposal occurs mainly in skeletal muscle
(70) via the glucose transporter (GLUT) 4, which is activated via the insulin signaling
pathway (71). Here insulin binds to the insulin receptor to activate its tyrosine kinase
activity. This phosphorylates IRS-1 on tyrosine residues allowing for the recruitment of the
p85/p110 PI3K to the plasma membrane. This generates PI(3,4,5)P3 from PI(4,5)P2, thereby
proefschrift Soeters.indb 16 4-8-2008 9:13:16
General introduction
17
Chapter 1
recruiting the 3’ phosphoinositide-dependent kinase-1 (PDK-1). PDK-1 phosphorylates and
activates both protein kinase B (AKT) and the atypical PKC λ/ζ (aPKCs) (71). PKCζ docks to
munc18c, resulting in enhanced GLUT4 translocation (72). Additionally, phosphorylation
of AS160, a protein containing a Rab GTPase-activating protein domain, is required for
the insulin induced translocation of GLUT4 to the plasma membrane (73). AS160 is a
downstream substrate of AKT.
It has been well established that peripheral (muscle) insulin mediated glucose uptake
decreases during short-term fasting (17;74-78). Interestingly, in animal studies it has
been shown that GLUT4 transcription decreases during fasting in adipose tissue (79;80).
However in skeletal muscle, GLUT4 mRNA is not altered after fasting in both animals and
humans (79;81). This indicates that during fasting posttranslational processes seem to
dictate the amount of GLUT4 recruitment (e.g. amount or phosphorylation of AKT). The
studies cited have not tried to explain lower insulin mediated glucose uptake after short-
term fasting, although Bergman et al suggest that the increased plasma FFA levels will
interfere with insulin mediated glucose uptake during fasting (82).
Fatty acids and insulin mediated glucose uptake
High plasma FFA are suggested to interfere with peripheral insulin mediated glucose
uptake, but the exact mechanisms are still not fully elucidated (83). It was demonstrated
in 1994 that increasing plasma FFA by an infusion of a lipid emulsion decreased insulin
mediated glucose uptake in healthy volunteers in a dose dependent fashion (84).
Interestingly, using the same study design, women were protected from FFA induced
insulin resistance compared to matched men, but again the exact mechanisms have not
been elucidated yet (85).
Various metabolites of FFA have been suggested to decrease insulin sensitivity in
skeletal muscle (43). The sphingolipid ceramide is one of these mediators. Increased
muscle ceramide concentrations have been reported in skeletal muscle of obese insulin
resistant subjects (86;87), while a negative correlation of muscle ceramide with insulin
sensitivity was found (87). Recently is shown that ceramide accumulates in muscle of
men at risk of developing type 2 diabetes (88). However, skeletal muscle ceramide levels
did not seem to play an important role in FFA associated insulin resistance in other
studies (89;90).
Intramyocellular ceramide content is mainly dependent on de novo synthesis from fatty
acids (91). In vitro studies showed that intracellular ceramide synthesis from palmitate is
one of the mechanisms by which palmitate interferes negatively with insulin-stimulated
phosphorylation of protein kinase B/AKT (AKT) (92;93). Furthermore, metabolites of
proefschrift Soeters.indb 17 4-8-2008 9:13:16
18
ceramide like complex glycosphingolipids (i.e. ganglioside GM3) may be involved in the
induction of insulin resistance (43;94;95).
One can hypothesize that fasting increases muscle ceramide levels, thereby decreasing
AKT phosphorylation and peripheral insulin mediated glucose uptake.
Additionally it is unknown whether the protection from FFA induced peripheral insulin
resistance in women can be explained by differences in muscle ceramide or FAT/CD36
levels.
In this thesis, we aimed to shed more light on the metabolic adaptation to fasting since
we are convinced that getting inside in the pathways which are involved in protecting the
body from energy depletion will provide us with answers on the metabolic adaptation to
overfeeding, i.e. the metabolic consequences of obesity.
proefschrift Soeters.indb 18 4-8-2008 9:13:16
General introduction
19
Chapter 1
Thesis OutlineSince women have higher circulating FFA and lower plasma glucose levels after short
term fasting we investigated in chapter 2 whether women would be relatively protected
from FFA induced insulin resistance due to lower ceramide levels in skeletal muscle. We
combined these measurements with analyses of the major FFA transporter FAT/CD36 in
skeletal muscle.
In chapter 3 we investigated whether muscle ceramide levels in skeletal muscle increase
during fasting. Hyperinsulinemic euglycemic clamps and muscle biopsies were performed.
We hypothesized muscle ceramide to increase during fasting, thereby reducing insulin
stimulated glucose uptake, possibly via decreased AKT phosphorylation.
In chapter 4 we explored the association of muscle long chain acylcarnitines with respect
to fatty acid metabolism and peripheral insulin sensitivity. We expected long-chain ACs
to increase during fasting and to correlate with whole body lipolysis, FAO rates and
peripheral insulin sensitivity after short-term fasting.
In chapter 5 we examined the quantative contributions of the L and D-hydroxybutyryl-
carnitine stereo isomers after short term fasting in lean healthy subjects. Additionally we
explored which biochemical synthetic route could be responsible; therefore additional
studies in liver en muscle homogenates of mice were performed.
In chapter 6 we explored the proposed insulin resistance in ketogenesis by studying
ketone body metabolism using stable isotopes in lean and obese subjects under equal
plasma insulin levels after short-term fasting. We expected ketogenesis to be equally
sensitive to insulin in obese versus lean subjects.
Intermittent fasting has been suggested to increase insulin stimulated glucose uptake
and thus insulin sensitivity although its mechanisms have not been elucidated. Therefore
in chapter 7 we tried to unravel some aspects of these beneficial effects by performing
hyperinsulinemic euglycemic clamps and muscle biopsies after eucaloric intermittent
fasting and a standard diet in crossover design in healthy volunteers. We focussed on
the effects of intermittent fasting on glucose, fat and protein metabolism using stable
isotopes.
proefschrift Soeters.indb 19 4-8-2008 9:13:16
20
Hypoinsulinemic hypoglycemia during fasting does usually not occur. However
clinicians sometimes encounter these cases. In chapter 8 we investigated 10 cases of
hypoinsulinemic hypoglycemia. EGP was measured as well as plasma acylcarnitines and
other intermediates of metabolism to rule out fatty acid oxidation disorders and disorders
in organic and amino acid metabolism.
Chapter 9 is a perspective describing the integrated metabolic response to fasting
regarding adaptive changes in lipid and glucose metabolism. The relevance and
physiological aspects of a mechanism that is needed for survival of the organism will be
discussed.
Reference List 1. Cahill J. Starvation in man. Clinics in Endocrinology and Metabolism 1976; 5(2):397-415.
2. Maltby JR. Fasting from midnight - the history behind the dogma. Best Practice & Research Clinical Anaesthesiology 2006; 20(3):363-378.
3. Sorkin JD, Muller DC, Fleg JL, Andres R. The Relation of Fasting and 2-h Postchallenge Plasma Glucose Concentrations to Mortality: Data from the Baltimore Longitudinal Study of Aging with a critical review of the literature. Diabetes Care 2005; 28(11):2626-2632.
4. Clore JN, Glickman PS, Helm ST, Nestler JE, Blackard WG. Accelerated decline in hepatic glucose production during fasting in normal women compared with men. Metabolism 1989; 38(11):1103-1107.
5. Klein S, Sakurai Y, Romijn JA, Carroll RM. Progressive alterations in lipid and glucose metabolism during short-term fasting in young adult men. Am J Physiol Endocrinol Metab 1993; 265(5):E801-E806.
6. Carlson MG, Snead WL, Campbell PJ. Fuel and energy metabolism in fasting humans. Am J Clin Nutr 1994; 60(1):29-36.
7. Klein S, Young VR, Blackburn GL, Bistrian BR, Wolfe RR. Palmitate and glycerol kinetics during brief starvation in normal weight young adult and elderly subjects. J Clin Invest 1986; 78(4):928-933.
8. Haymond MW, Karl IE, Clarke WL, Pagliara AS, Santiago JV. Differences in circulating gluconeogenic substrates during short-term fasting in men, women, and children. Metabolism 1982; 31(1):33-42.
9. Merimee TJ, Fineberg SE. Homeostasis during fasting. II. Hormone substrate differences between men and women. J Clin Endocrinol Metab 1973; 37(5):698-702.
10. Fryburg DA, Barrett EJ, Louard RJ, Gelfand RA. Effect of starvation on human muscle protein metabolism and its response to insulin. Am J Physiol Endocrinol Metab 1990; 259(4):E477-E482.
11. Tsalikian E, Howard C, Gerich JE, Haymond MW. Increased leucine flux in short-term fasted human subjects: evidence for increased proteolysis. Am J Physiol Endocrinol Metab 1984; 247(3):E323-E327.
12. Landau BR, Wahren J, Chandramouli V, Schumann WC, Ekberg K, Kalhan SC. Contributions of gluconeogenesis to glucose production in the fasted state. J Clin Invest 1996; 98(2):378-385.
13. Randle PJ, Garland PB, Hales CN, Newsholme EA. The glucose fatty-acud cycle: its role in insulin sensitivity and the disturbances of diabetes mellitus. The Lancet 1963; 281(7285):785-789.
proefschrift Soeters.indb 20 4-8-2008 9:13:16
General introduction
21
Chapter 1
14. Romijn JA, Godfried MH, Hommes MJT, Endert E, Sauerwein HP. Decreased glucose oxidation during short-term starvation. Metabolism 1990; 39(5):525-530.
15. Norrelund H, Nielsen S, Christiansen JS, Jorgensen JO, Moller N. Modulation of basal glucose metabolism and insulin sensitivity by growth hormone and free fatty acids during short-term fasting. Eur J Endocrinol 2004; 150(6):779-787.
16. Stannard SR, Thompson MW, Fairbairn K, Huard B, Sachinwalla T, Thompson CH. Fasting for 72 h increases intramyocellular lipid content in nondiabetic, physically fit men. Am J Physiol Endocrinol Metab 2002; 283(6):E1185-E1191.
17. Bjorkman O, Eriksson LS. Influence of a 60-hour fast on insulin-mediated splanchnic and peripheral glucose metabolism in humans. J Clin Invest 1985; 76(1):87-92.
18. Nair KS, Woolf PD, Welle SL, Matthews DE. Leucine, glucose, and energy metabolism after 3 days of fasting in healthy human subjects. Am J Clin Nutr 1987; 46(4):557-562.
19. Stryer L. Food intake and starvation induce metabolic change. In: Tymoczko, Berg, editors. Biochemistry International Edition. Freeman, 2005: 854-858.
20. Rigaud D, Hassid J, Meulemans A, Poupard AT, Boulier A. A paradoxical increase in resting energy expenditure in malnourished patients near death: the king penguin syndrome. Am J Clin Nutr 2000; 72(2):355-360.
21. Roky R, Houti I, Moussamih S, Qotbi S, Aadil N. Physiological and Chronobiological Changes during Ramadan Intermittent Fasting. Annals of Nutrition & Metabolism 2004; 48(4):296-303.
22. Halberg N, Henriksen M, Soderhamn N et al. Effect of intermittent fasting and refeeding on insulin action in healthy men. J Appl Physiol 2005; 99(6):2128-2136.
23. Speakman JR. Thrifty genes for obesity and the metabolic syndrome--time to call off the search? Diab Vasc Dis Res 2006; 3(1):7-11.
24. Bouchard C. The biological predisposition to obesity: beyond the thrifty genotype scenario. Int J Obes 2007; 31(9):1337-1339.
25. Varady KA, Hellerstein MK. Alternate-day fasting and chronic disease prevention: a review of human and animal trials. Am J Clin Nutr 2007; 86(1):7-13.
26. Heilbronn LK, Civitarese AE, Bogacka I, Smith SR, Hulver M, Ravussin E. Glucose tolerance and skeletal muscle gene expression in response to alternate day fasting. Obes Res 2005; 13(3):574-581.
27. Heilbronn LK, Smith SR, Martin CK, Anton SD, Ravussin E. Alternate-day fasting in nonobese subjects: effects on body weight, body composition, and energy metabolism. Am J Clin Nutr 2005; 81(1):69-73.
28. Hojlund K, Wildner-Christensen M, Eshoj O et al. Reference intervals for glucose, {beta}-cell polypeptides, and counterregulatory factors during prolonged fasting. Am J Physiol Endocrinol Metab 2001; 280(1):E50-E58.
29. Ryden M, Jocken J, van Harmelen V et al. Comparative studies of the role of hormone-sensitive lipase and adipose triglyceride lipase in human fat cell lipolysis. Am J Physiol Endocrinol Metab 2007; 292(6):E1847-E1855.
30. Kraemer FB, Shen WJ. Hormone-sensitive lipase: control of intracellular tri-(di-)acylglycerol and cholesteryl ester hydrolysis. J Lipid Res 2002; 43(10):1585-1594.
31. Sztalryd C, Xu G, Dorward H et al. Perilipin A is essential for the translocation of hormone-sensitive lipase during lipolytic activation. J Cell Biol 2003; 161(6):1093-1103.
32. Miyoshi H, Souza SC, Zhang HH et al. Perilipin promotes hormone-sensitive lipase-mediated adipocyte lipolysis via phosphorylation-dependent and -independent mechanisms. J Biol Chem 2006; 281(23):15837-15844.
33. Zimmermann R, Strauss JG, Haemmerle G et al. Fat Mobilization in Adipose Tissue Is Promoted by Adipose Triglyceride Lipase. Science 2004; 306(5700):1383-1386.
34. Spriet LL, Tunstall RJ, Watt MJ, Mehan KA, Hargreaves M, Cameron-Smith D. Pyruvate dehydrogenase activation and kinase expression in human skeletal muscle during fasting. J Appl Physiol 2004; 96(6):2082-2087.
proefschrift Soeters.indb 21 4-8-2008 9:13:16
22
35. Osler ME, Zierath JR. Minireview: Adenosine 5‘-Monophosphate-Activated Protein Kinase Regulation of Fatty Acid Oxidation in Skeletal Muscle. Endocrinology 2008; 149(3):935-941.
36. Wu P, Inskeep K, Bowker-Kinley MM, Popov KM, Harris RA. Mechanism responsible for inactivation of skeletal muscle pyruvate dehydrogenase complex in starvation and diabetes. Diabetes 1999; 48(8):1593-1599.
37. Boden G, Shulman GI. Free fatty acids in obesity and type 2 diabetes: defining their role in the development of insulin resistance and beta-cell dysfunction. Eur J Clin Invest 2002; 32 Suppl 3:14-23.
38. Jensen MD, Ekberg K, Landau BR. Lipid metabolism during fasting. Am J Physiol Endocrinol Metab 2001; 281(4):E789-E793.
39. Chabowski A, Gorski J, Luiken JJ, Glatz JF, Bonen A. Evidence for concerted action of FAT/CD36 and FABPpm to increase fatty acid transport across the plasma membrane. Prostaglandins Leukot Essent Fatty Acids 2007; 77(5-6):345-353.
40. Stephens FB, Constantin-Teodosiu D, Greenhaff PL. New insights concerning the role of carnitine in the regulation of fuel metabolism in skeletal muscle. J Physiol 2007; 581(2):431-444.
41. Pollitt RJ. Disorders of mitochondrial long-chain fatty acid oxidation. J Inherit Metab Dis 1995; 18(4):473-490.
42. Wanders RJ, Vreken P, den Boer ME, Wijburg FA, van Gennip AH, IJlst L. Disorders of mitochondrial fatty acyl-CoA beta-oxidation. J Inherit Metab Dis 1999; 22(4):442-487.
43. Holland WL, Knotts TA, Chavez JA, Wang LP, Hoehn KL, Summers SA. Lipid Mediators of Insulin Resistance. Nutrition Reviews 2007; 65:39-46.
44. An J, Muoio DM, Shiota M et al. Hepatic expression of malonyl-CoA decarboxylase reverses muscle, liver and whole-animal insulin resistance. Nat Med 2004; 10(3):268-274.
45. Koves TR, Ussher JR, Noland RC et al. Mitochondrial overload and incomplete fatty acid oxidation contribute to skeletal muscle insulin resistance. Cell Metab 2008; 7(1):45-56.
46. Muoio DM, Koves TR. Lipid-induced metabolic dysfunction in skeletal muscle. Novartis Found Symp 2007; 286:24-38; discussion 38-46, 162-3,;%196-203.:24-38.
47. Hoppel CL, Genuth SM. Carnitine metabolism in normal-weight and obese human subjects during fasting. Am J Physiol 1980; 238(5):E409-E415.
48. Frohlich J, Seccombe DW, Hahn P, Dodek P, Hynie I. Effect of fasting on free and esterified carnitine levels in human serum and urine: Correlation with serum levels of free fatty acids and [beta]-hydroxybutyrate. Metabolism 1978; 27(5):555-561.
49. Costa CCG, De Almeida IT, Jakobs CORN, Poll-The B-T, DuranURAN M. Dynamic Changes of Plasma Acylcarnitine Levels Induced by Fasting and Sunflower Oil Challenge Test in Children. [Miscellaneous Article]. Pediatric Research 1999; 46(4):440.
50. Balasse EO. Kinetics of ketone body metabolism in fasting humans. Metabolism 1979; 28(1):41-50.
51. Owen OE, Morgan AP, Kemp HG, Sullivan JM, Herrera MG, Cahill GF, Jr. Brain metabolism during fasting. J Clin Invest 1967; 46(10):1589-1595.
52. Nadal A, Marrero PF, Haro D. Down-regulation of the mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase gene by insulin: the role of the forkhead transcription factor FKHRL1. Biochem J 2002; 366(Pt 1):289-297.
53. Singh BM, Krentz AJ, Nattrass M. Insulin resistance in the regulation of lipolysis and ketone body metabolism in non-insulin dependent diabetes is apparent at very low insulin concentrations. Diabetes Research and Clinical Practice 1993; 20(1):55-62.
54. Hack A, Busch V, Pascher B et al. Monitoring of ketogenic diet for carnitine metabolites by subcutaneous microdialysis. Pediatr Res 2006; 60(1):93-96.
55. Swiatek KR, Dombrowski GJ, Chao KL, Chao HL. Metabolism of - and -3-hydroxybutyrate by rat liver during development. Biochemical Medicine 1981; 25(2):160-167.
proefschrift Soeters.indb 22 4-8-2008 9:13:16
General introduction
23
Chapter 1
56. Hildebrandt LA, Spennetta T, Elson C, Shrago E. Utilization and preferred metabolic pathway of ketone bodies for lipid synthesis by isolated rat hepatoma cells. Am J Physiol Cell Physiol 1995; 269(1):C22-C27.
57. Hildebrandt L, Spennetta T, Ackerman R, Elson C, Shrago E. Acyl CoA and Lipid Synthesis from Ketone Bodies by the Extramitochondrial Fraction of Hepatoma Tissue. Biochemical and Biophysical Research Communications 1996; 225(1):307-312.
58. Molven A, Matre GE, Duran M et al. Familial Hyperinsulinemic Hypoglycemia Caused by a Defect in the SCHAD Enzyme of Mitochondrial Fatty Acid Oxidation. Diabetes 2004; 53(1):221-227.
59. Fery F. Role of hepatic glucose production and glucose uptake in the pathogenesis of fasting hyperglycemia in type 2 diabetes: normalization of glucose kinetics by short-term fasting. J Clin Endocrinol Metab 1994; 78(3):536-542.
60. Rothman DL, Magnusson I, Katz LD, Shulman RG, Shulman GI. Quantitation of hepatic glycogenolysis and gluconeogenesis in fasting humans with 13C NMR. Science 1991; 254(5031):573-576.
61. Cano N. Bench-to-bedside review: glucose production from the kidney. Crit Care 2002; 6(4):317-321.
62. Mittendorfer B, Horowitz JF, Klein S. Gender differences in lipid and glucose kinetics during short-term fasting. Am J Physiol Endocrinol Metab 2001; 281(6):E1333-E1339.
63. Service FJ. Hypoglycemic Disorders. N Engl J Med 1995; 332(17):1144-1152.
64. Service FJ, Natt N. The Prolonged Fast. J Clin Endocrinol Metab 2000; 85(11):3973-3974.
65. Corssmit EPM, Romijn JA, Sauerwein HP. Review article: Regulation of glucose production with special attention to nonclassical regulatory mechanisms: A review. Metabolism 2001; 50(7):742-755.
66. Boden G. Effects of free fatty acids on gluconeogenesis and glycogenolysis. Life Sci 2003; 72(9):977-988.
67. Boden G. Interaction between free fatty acids and glucose metabolism. Curr Opin Clin Nutr Metab Care 2002; 5(5):545-549.
68. Chen X, Iqbal N, Boden G. The effects of free fatty acids on gluconeogenesis and glycogenolysis in normal subjects. J Clin Invest 1999; 103(3):365-372.
69. Fery F, Plat L, Melot C, Balasse EO. Role of fat-derived substrates in the regulation of gluconeogenesis during fasting. Am J Physiol Endocrinol Metab 1996; 270(5):E822-E830.
70. DeFronzo RA, Jacot E, Jequier E, Maeder E, Wahren J, Felber JP. The effect of insulin on the disposal of intravenous glucose. Results from indirect calorimetry and hepatic and femoral venous catheterization. Diabetes 1981; 30(12):1000-1007.
71. Khan AH, Pessin JE. Insulin regulation of glucose uptake: a complex interplay of intracellular signalling pathways. Diabetologia 2002; 45(11):1475-1483.
72. Hodgkinson CP, Mander A, Sale GJ. Protein kinase-zeta interacts with munc18c: role in GLUT4 trafficking. Diabetologia 2005; 48(8):1627-1636.
73. Karlsson HKR, Zierath JR, Kane S, Krook A, Lienhard GE, Wallberg-Henriksson H. Insulin-Stimulated Phosphorylation of the Akt Substrate AS160 Is Impaired in Skeletal Muscle of Type 2 Diabetic Subjects. Diabetes 2005; 54(6):1692-1697.
74. Anderson JW, Herman RH. Effects of carbohydrate restriction on glucose tolerance of normal men and reactive hypoglycemic patients. Am J Clin Nutr 1975; 28(7):748-755.
75. Anderson JW, Herman RH. Effect of fasting, caloric restriction, and refeeding on glucose tolerance of normal men. Am J Clin Nutr 1972; 25(1):41-52.
76. Bergman BC, Cornier MA, Horton TJ, Bessesen D. Effects Of Fasting On Insulin Action And Glucose Kinetics In Lean And Obese Men And Women. Am J Physiol Endocrinol Metab 2007;00613.
77. Newman WP, Brodows RG. Insulin action during acute starvation: evidence for selective insulin resistance in normal man. Metabolism 1983; 32(6):590-596.
proefschrift Soeters.indb 23 4-8-2008 9:13:16
24
78. Mansell PI, Macdonald IA. The effect of starvation on insulin-induced glucose disposal and thermogenesis in humans. Metabolism 1990; 39(5):502-510.
79. Ezaki O. Regulatory Elements in the Insulin-Responsive Glucose Transporter (GLUT4) Gene. Biochemical and Biophysical Research Communications 1997; 241(1):1-6.
80. Sivitz WI, DeSautel SL, Kayano T, Bell GI, Pessin JE. Regulation of glucose transporter messenger RNA in insulin-deficient states. Nature 1989; 340(6228):72-74.
81. Pilegaard H, Saltin B, Neufer PD. Effect of Short-Term Fasting and Refeeding on Transcriptional Regulation of Metabolic Genes in Human Skeletal Muscle. Diabetes 2003; 52(3):657-662.
82. Bergman BC, Cornier MA, Horton TJ, Bessesen D. Effects Of Fasting On Insulin Action And Glucose Kinetics In Lean And Obese Men And Women. Am J Physiol Endocrinol Metab 2007;00613.
83. Charron MJ, Kahn BB. Divergent molecular mechanisms for insulin-resistant glucose transport in muscle and adipose cells in vivo. J Biol Chem 1990; 265(14):7994-8000.
84. Boden G, Chen X, Ruiz J, White JV, Rossetti L. Mechanisms of fatty acid-induced inhibition of glucose uptake. J Clin Invest 1994; 93(6):2438-2446.
85. Frias JP, Macaraeg GB, Ofrecio J, Yu JG, Olefsky JM, Kruszynska YT. Decreased susceptibility to fatty acid-induced peripheral tissue insulin resistance in women. Diabetes 2001; 50(6):1344-1350.
86. Adams JM, Pratipanawatr T, Berria R et al. Ceramide content is increased in skeletal muscle from obese insulin-resistant humans. Diabetes 2004; 53(1):25-31.
87. Straczkowski M, Kowalska I, Nikolajuk A et al. Relationship between insulin sensitivity and sphingomyelin signaling pathway in human skeletal muscle. Diabetes 2004; 53(5):1215-1221.
88. Straczkowski M, Kowalska I, Baranowski M et al. Increased skeletal muscle ceramide level in men at risk of developing type 2 diabetes. Diabetologia 2007; 50(11):2366-2373.
89. Serlie MJ, Allick G, Groener JE et al. Chronic Treatment With Pioglitazone Does Not Protect Obese Patients With Diabetes Mellitus Type II from Free Fatty Acid-Induced Insulin Resistance. J Clin Endocrinol Metab 2006; .
90. Serlie MJ, Meijer AJ, Groener JE et al. Short-Term Manipulation of Plasma Free Fatty Acids Does Not Change Skeletal Muscle Concentrations of Ceramide and Glucosylceramide in Lean and Overweight Subjects. J Clin Endocrinol Metab 2007; 92(4):1524-1529.
91. Summers SA. Ceramides in insulin resistance and lipotoxicity. Prog Lipid Res 2006; 45(1):42-72.
92. Powell DJ, Turban S, Gray A, Hajduch E, Hundal HS. Intracellular ceramide synthesis and protein kinase Czeta activation play an essential role in palmitate-induced insulin resistance in rat L6 skeletal muscle cells. Biochem J 2004; 382(Pt 2):619-629.
93. Schmitz-Peiffer C, Craig DL, Biden TJ. Ceramide generation is sufficient to account for the inhibition of the insulin-stimulated PKB pathway in C2C12 skeletal muscle cells pretreated with palmitate. J Biol Chem 1999; 274(34):24202-24210.
94. Aerts JM, Ottenhoff R, Powlson AS et al. Pharmacological Inhibition of Glucosylceramide Synthase Enhances Insulin Sensitivity. Diabetes 2007; 56(5):1341-1349.
95. Tagami S, Inokuchi Ji, Kabayama K et al. Ganglioside GM3 Participates in the Pathological Conditions of Insulin Resistance. J Biol Chem 2002; 277(5):3085-3092.
proefschrift Soeters.indb 24 4-8-2008 9:13:16
proefschrift Soeters.indb 25 4-8-2008 9:13:16
proefschrift Soeters.indb 26 4-8-2008 9:13:17
Gender related differences in the metabolic response to fasting2Maarten R. Soeters1, Hans P. Sauerwein1,
Johanna E. Groener2, Johannes M. Aerts2,
Mariëtte T. Ackermans3, Jan F.C. Glatz4,
Eric Fliers1 and Mireille J. Serlie1
1Department of Endocrinology and Metabolism 2Department of Medical Biochemistry, 3Department of Clinical Chemistry, Laboratory
of Endocrinology, Academic Medical Center,
University of Amsterdam, Amsterdam, the
Netherlands, 4Department of Molecular
Genetics, Maastricht University, Maastricht, the
Netherlands
J Clin Endocrinol Metab 2007;
92(9):3646-3652.
proefschrift Soeters.indb 27 4-8-2008 9:13:17
28
Abstract
Context: Free fatty acids (FFA) may induce insulin resistance via synthesis of intramyocellular
ceramide. During fasting, women have lower plasma glucose levels than men despite
higher plasma FFA, suggesting protection from FFA-induced insulin resistance.
Objective: We studied whether the relative protection from FFA-induced insulin resistance
during fasting in women, is associated with lower muscle ceramide concentrations
compared to men.
Main Outcome Measures and Design: After a 38 h fast, measurements of glucose and
lipid fluxes and muscle ceramide and fatty acid translocase/CD36 were performed before
and after a hyperinsulinaemic euglycaemic clamp.
Results: Plasma glucose levels were significantly lower in women than men with a
trend for a lower endogenous glucose production in women, while FFA and lipolysis
were significantly higher. Insulin-mediated peripheral glucose uptake was not different
between sexes. There was no gender difference in muscle ceramide in the basal state and
ceramide did not correlate with peripheral glucose uptake. Muscle fatty acid translocase/
CD36 was not different between sexes in the basal state and during the clamp.
Conclusion: After 38 h of fasting, plasma FFA were higher and plasma glucose was
lower in women compared to men. The higher plasma FFA did not result in differences
in peripheral insulin sensitivity, possibly because of similar muscle ceramide and fatty acid
translocase/CD36 levels in men and women. We suggest that during fasting, women
are relatively protected from FFA-induced insulin resistance by preventing myocellular
accumulation of ceramide.
proefschrift Soeters.indb 28 4-8-2008 9:13:17
Gender differences in fasting
29
Chapter 2
Introduction
The adaptive metabolic response to short-term fasting consists of integrated metabolic
alterations that guarantee substrate availability for energy production and prevent
hypoglycaemia. During fasting, plasma insulin levels are low and plasma concentrations
of catecholamines, glucagon and growth hormone are increased, resulting in increased
lipolysis and thus high plasma free fatty acid (FFA) concentrations (1-3).
It has been known for a long time that women have lower plasma glucose and higher
plasma FFA concentrations than men after short-term fasting (1;4-6). However, despite
extensive research on gender related distinctions in glucose and lipid metabolism, these
differences in plasma glucose and FFA concentrations have not been explained in full
detail so far (7-16).
In women, the combination of lower plasma glucose levels on one hand, and higher
plasma FFA levels on the other hand, is intriguing, since it is generally accepted that
high plasma FFA levels increase endogenous glucose production (EGP) and decrease
peripheral glucose uptake (17;18). Consistently, it has been shown that women are
relatively protected from FFA-induced insulin resistance (12;16).
The exact underlying mechanisms by which FFA interfere with insulin signalling have
not yet been unravelled completely. One potential mechanism may involve the de novo
synthesis of ceramide from palmitate, because intramyocellular ceramide was found to
be increased in obese insulin-resistant patients and correlated with whole body insulin
sensitivity (19;20). Moreover, in vitro studies showed that intracellular ceramide synthesis
from palmitate was found to be one of the mechanisms by which palmitate interferes
negatively with insulin-stimulated phosphorylation of protein kinase B (PKB) (21;22).
Furthermore, metabolites from ceramide (i.e. glycosphingolipids, like glucosylceramide)
might be involved in the induction of insulin resistance (23;24). Since intramyocellular
ceramide concentration correlates positively with plasma FFA levels, it might be expected
that the increased levels of plasma FFA in women result in higher muscle ceramide levels
(20). However, this would contradict with the reported relative protection from FFA-
induced insulin resistance in women.
Two mechanisms may explain this relative insensitivity to increased plasma FFA levels:
firstly, lower myocellular uptake of plasma FFA and secondly, differences in muscle
fatty-acid handling. Cellular uptake of plasma FFA occurs by protein-mediated transport
and via flip-flop of protonated fatty acids (25-27), depending on transmembrane
concentration gradients and intracellular fatty acid metabolism (25;27-29). Fatty acid
proefschrift Soeters.indb 29 4-8-2008 9:13:17
30
translocase (FAT)/CD36 is the main protein involved in muscle fatty acid uptake (26).
Fatty-acid handling involves storage into complex lipids or oxidation of fatty acyl-CoAs.
To the best of our knowledge, gender differences in muscle (glyco)sphingolipids and the
fatty acid transporter CD36 content after short-term fasting in relation to glucose and
lipid metabolism have not been studied before.
In this study, we measured glucose and lipid fluxes after short-term fasting in healthy
lean men and women in the basal state and during a hyperinsulinaemic euglycaemic
clamp (stable isotope technique). Furthermore, we assessed total muscle content of
ceramide, glucosylceramide and the lipid binding protein FAT/CD36 in the basal state
and during the clamp. We hypothesized that the relative protection from FFA-induced
insulin resistance during fasting in women results from lower muscle ceramide or
glucosylceramide levels due to lower muscle FFA uptake, subsequently resulting in higher
muscle glucose uptake with lower plasma glucose concentrations.
Subjects and methods
SubjectsTen male and 10 female subjects were recruited via advertisements in local magazines.
Criteria for inclusion were 1) absence of a family history of diabetes; 2) age 18–35 yr; 3)
Caucasian race; 4) BMI 20-25 kg/m2; 5) no excessive sport activities, i.e. < 3 times per
week; and 6) no medication. Women were studied during the follicular phase of the
menstrual cycle. Subjects were in self-reported good health, confirmed by medical history
and physical examination. Written informed consent was obtained from all subjects after
explanation of purposes, nature, and potential risks of the study. The study was approved
by the Medical Ethical Committee of the Academic Medical Center of the University of
Amsterdam.
Experimental protocolFor three days before the fasting period, all volunteers consumed a weight-maintaining
diet containing at least 250 g of carbohydrates per day. Then, the subjects were fasting
from 2000 h two days before the start of the study until the end of the study. Volunteers
were admitted to the metabolic unit of the Academic Medical Center at 0730 h. Subjects
were studied in the supine position and were allowed to drink water only.
A catheter was inserted into an antecubital vein for infusion of stable isotope tracers,
insulin and glucose. Another catheter was inserted retrogradely into a contralateral hand
proefschrift Soeters.indb 30 4-8-2008 9:13:17
Gender differences in fasting
31
Chapter 2
vein and kept in a thermo-regulated (60°C) plexiglas box for sampling of arterialized
venous blood. In all studies, saline was infused as NaCl 0.9% at a rate of 50 mL/h to keep
the catheters patent. [6,6-2H2]glucose and [1,1,2,3,3-2H5]glycerol were used as tracers
(>99% enriched; Cambridge Isotopes, Andover, USA).
To study total triglyceride hydrolysis, we used [1,1,2,3,3-2H5]glycerol . This may result
in underestimation of FFA release (in contrast to a fatty acid tracer). However, curves of
glycerol and fatty acid tracers are very similar during fasting (2), although the latter is
preferred to quantify adipose tissue lipolysis (30).
At T = 0 h (0800 h), blood samples were drawn for determination of background
enrichments and a primed continuous infusion of both isotopes was started: [6,6-2H2]
glucose (prime, 8.8 μmol/kg; continuous, 0.11 μmol/kg·min) and [1,1,2,3,3-2H5]glycerol
(prime, 1.6 μmol/kg; continuous, 0.11 μmol/kg·min) and continued until the end of the
study. After an equilibration period of two hours (38 h of fasting), 3 blood samples were
drawn for glucose and glycerol enrichments and 1 for glucoregulatory hormones, FFA
and adiponectin. Thereafter (T = 3 h), infusions of insulin (60mU/m2·min) (Actrapid 100
IU/ml; Novo Nordisk Farma B.V., Alphen aan den Rijn, the Netherlands) and glucose 20%
(to maintain a plasma glucose level of 5 mmol/L) were started. [6,6-2H2]glucose was
added to the 20% glucose solution to achieve glucose enrichments of 1% to approximate
the values for enrichment reached in plasma and thereby minimizing changes in isotopic
enrichment due to changes in the infusion rate of exogenous glucose. Plasma glucose
levels were measured every 5 minutes at the bedside. At T = 8 h, 5 blood samples
were drawn at 5 minute intervals for determination of glucose and glycerol enrichments.
Another blood sample was drawn for determination of glucoregulatory hormones, FFA
and adiponectin.
Body composition and indirect calorimetryBody composition was measured with bioelectrical impedance analysis (Maltron BF 906,
Rayleigh, UK). Oxygen consumption (VO2) and CO2 production (VCO2) were measured
continuously during the final 20 min of both the basal state and the hyperinsulinaemic
euglycaemic clamp by indirect calorimetry using a ventilated hood system (Sensormedics
model 2900; Sensormedics, Anaheim, USA).
Muscle biopsyMuscle biopsies were performed to assess muscle content of ceramide, glucosylceramide
and FAT/CD36 at the end of both the basal state and the hyperinsulinaemic euglycaemic
clamp. The muscle biopsy was performed under local anaesthesia (lidocaine 20 mg/ml;
proefschrift Soeters.indb 31 4-8-2008 9:13:17
32
Fresenius Kabi; Den Bosch, the Netherlands) using a Pro-MagTM I Biopsy Needle (MDTECH,
Gainesville, USA). Biopsy specimens were quickly washed in a buffer (NaCl 0.9%/Hepes
28.3gr/L) to remove blood, inspected for fat or fascia content, dried on gauze swabs and
subsequently stored in liquid nitrogen until analysis.
Glucose and lipid metabolism measurementsPlasma glucose concentrations were measured with the glucose oxidase method using
a Beckman Glucose Analyzer 2 (Beckman, Palo Alto, USA, intra-assay variation 2-3%).
Plasma FFA concentrations were determined with an enzymatic colorimetric method
(NEFA-C test kit, Wako Chemicals GmbH, Neuss, Germany): intra-assay variation 1%;
inter-assay variation: 4-15%; detection limit: 0.02 mmol/L. [6,6-2H2]glucose enrichment
was measured as described earlier (31). [6,6-2H2]glucose enrichment (tracer/tracee
ratio) intra-assay variation: 0.5-1%; inter-assay variation 1%; detection limit: 0.04%.
[1,1,2,3,3-2H5]glycerol enrichment was determined as described earlier (32). Intra-assay
variation glycerol: 1-3%, [1,1,2,3,3-2H5]glycerol: 4%; inter-assay variation glycerol: 2-3%;
[1,1,2,3,3-2H5]glycerol: 7%.
Glucoregulatory hormones and adiponectinInsulin and cortisol were determined on an 2000 system (Diagnostic Products Corporation,
Los Angeles, USA). Insulin was measured with a chemiluminescent immunometric
assay, intra-assay variation: 3-6%, inter-assay variation: 4-6%, detection limit: 15pmol/L.
Cortisol was measured with a chemiluminescent immunoassay, intra-assay variation:
7-8%, inter-assay variation: 7-8%, detection limit: 50nmol/L. Glucagon was determined
with the Linco 125I radioimmunoassay (St. Charles, USA), intra-assay variation: 9-10%,
inter-assay variation: 5-7% and detection limit: 15ng/L. Norepinephrine and epinephrine
were determined with an in-house HPLC method. Intra-assay variation norepinephrine:
2%; epinephrine 9%; inter-assay variation norepinephrine: 10%; epinephrine: 14 -18%;
detection limit: 0.05 nmol/L. Adiponectin was determined by a radioimmunoassay (Linco,
St. Charles, USA)(Intra-assay variation: 2-7%; inter-assay variation: 16-17%; detection
limit: 1 ng/mL).
Ceramide and glucosylceramide measurementsCeramide and glucosylceramide in muscle biopsies were measured with a high
performance liquid chromatography method as described (33). Muscle biopsies were
weighed and homogenized in 300 μl water by sonification. 50 μl muscle homogenates
proefschrift Soeters.indb 32 4-8-2008 9:13:17
Gender differences in fasting
33
Chapter 2
were used. All samples were run in duplicate and in every run 2 reference samples were
included. CV: Inter-assay 4%, intra-assay < 14 %.
FAT/CD36 measurementsMuscle biopsies were homogenized four times 5 seconds (Ultra-Turrax®, Ika-Werke,
Staufen, Germany) in 10 volumes of cold (4º C) buffer containing 250 mM sucrose, 2 mM
Na-EDTA and 10 mM Tris at pH 7.4, followed by four times 5 sec ultrasonic treatment
(Ultrasonic Processor, Hielsher GmbH, Teltow, Germany). Total protein was determined
with the bicinchonic acid method (Pierce, Rockford IL, USA). All samples were diluted
to the same protein concentration and mixed with sample buffer (4:1, vol/vol) before
being subjected to SDS-polyacrylamide gel electrophoresis and immunoblotting exactly
as described before (34).
The CD36 antigen-antibody band at 88 kDa was visualized with a chemiluminescence
substrate (ECL, Amersham Biosciences UK Ltd, Buckinghamshire, England) and quantified
using Quantity One software (Bio-Rad, Hercules CA, USA). Rat heart and liver whole
homogenates were used as positive and negative controls, respectively.
Calculations and statisticsEndogenous glucose production (EGP) and peripheral glucose uptake (rate of
disappearance/Rd) were calculated using the modified forms of the Steele Equations
as described previously (35;36). EGP and Rd were expressed as μmol/kg·min. Glucose
metabolic clearance rates (MCR) were calculated as MCR = Ra / [glucose]. Lipolysis
(glycerol turnover) was calculated by using formulas for steady state kinetics adapted
for stable isotopes (32). Lipolysis was expressed as μmol/kg·min and as μmol/kcal as
proposed by Koutsari et al (30). Resting energy expenditure (REE) and glucose and fat
oxidation rates were calculated from O2 consumption and CO2 production as reported
previously (37).
All data were analyzed with non parametric tests. Comparisons between groups (at
T = 2 and 8 h) were performed using the Mann-Whitney U test. Comparisons within
groups (between T = 2 and 8 h) were performed with the Wilcoxon Signed Rank test.
Correlations were expressed as Spearman’s rank correlation coefficient (ρ). The SPSS
statistical software program version 12.0.1 (SPSS Inc, Chicago, IL) was used for statistical
analysis. Data are presented as median [minimum -maximum].
proefschrift Soeters.indb 33 4-8-2008 9:13:17
34
Results
Anthropometric characteristicsMen and women did not differ in age or BMI (Table 1). Weight and percentage lean
body mass were higher and percentage fat mass was lower in men (Table 1).
Resting energy expenditure, glucose and lipid kineticsTotal REE was higher in men than in women, both in the basal state and during the
hyperinsulinaemic clamp (Table 2a). Rates of glucose and fat oxidation did not differ
between men and women in the basal state and during the clamp (Table 2a). In the
female group, plasma glucose concentrations were significantly lower after 38 h of
fasting compared to the male group (Table 2b). Basal EGP (and Rd) tended to be lower
in women (Table 2b). Insulin-mediated peripheral glucose uptake during the clamp (Rd)
did not differ significantly between men and women (Table 2b). MCR in the basal state
was not different between women and men (2.0 [1.8 – 3.2] ml/kg•min vs. 2.1 [1.7
– 2.4] ml/kg•min respectively, P = 1). Basal plasma FFA were significantly higher in
Table 1 Clinical characteristics of male and female subjects.
men (n = 10) women (n = 10) p
Age(years) 21.3 [18.9 - 25.1] 22.2 [19.1 - 28.9] 0.3
Weight (kg) 74.6 [65.5 - 89.0] 63.5 [54.5 - 72.0] 0.001
Heigth (cm) 187 [179 - 195] 169 [160 - 177] 0.4
BMI (kg/m2) 21.5 [19.2 - 24.7] 22.9 [18.7 - 24.2] 0.4
Lean body mass (%) 89 [77 - 91] 75 [70 - 89] 0.002
Fat mass (%) 11 [9 - 23] 25 [11 - 30] 0.002
Data are presented as median [min - max]. BMI = body mass index.
Table 2 REE and oxidation rates for glucose and fat.
basal state hyperinsulinaemic euglycaemic clamp
men (n = 10) women (n = 10) P men (n = 10) women (n = 10) P
REE (kcal/day) 1995 [1840 - 2356] 1511 [884 - 1833] 0.001 1992 [1685 - 2177] 1467 [1150 - 1842] 0.001
Glucose oxidation (μmol/kg•min) 5.0 [0 - 18.3] 3.9 [0 - 7.8] 0.5 20 [8.9 - 33.3] 16.1 [11.7 - 28.3] 0.4
Glucose oxidation (μmol/kg lbm•min) 5.6 [0 - 20.6] 5.0 [0 - 10.6] 0.8 22.2 [11.1 - 37.8] 21.7 [13.9 - 38.3] 0.9
Fat oxidation (μmol/kg•min) 2.0 [0.9 - 2.8] 1.7 [1.3 - 2.1] 0.3 0.7 [0 - 1.5] 0.5 [ 0 - 1.0] 0.4
Fat oxidation (μmol/kg lbm•min) 2.3 [1.0 - 3.0] 2.2 [1.9 - 2.9] 0.9 0.8 [0 - 1.7] 0.6 [0 - 1.3] 0.7
Data are presented as median [minimum - maximum]. REE = resting energy expenditure, lbm = lean body mass.
proefschrift Soeters.indb 34 4-8-2008 9:13:17
Gender differences in fasting
35
Chapter 2
women compared to men, but were equally suppressed during the hyperinsulinaemic
euglycaemic clamp in both groups (Table 2b). Lipolysis expressed as μmol/kg·min was
not different between men and women in the basal state and during the clamp, but
when expressed in μmol/kcal it was shown that women had significant higher lipolysis
rates in the basal state, but not during the clamp (Table 2b).
Glucoregulatory hormones and adiponectinPlasma insulin, cortisol, glucagon and norepinephrine levels were not different between
sexes in the basal state and during the clamp (Table 3). Plasma epinephrine was significantly
lower in females than males during the hyperinsulinaemic euglycaemic clamp (Table 3).
Adiponectin was significantly higher in females in both the basal state and during the
clamp (Table 3). Adiponectin decreased significantly from baseline during the clamp,
though no gender difference in relative decrease was observed (data not shown).
Ceramide and glucosylceramide measurementsMuscle ceramide concentrations in the basal state were not different between sexes
(Figure 1). There was a trend to lower muscle ceramide levels in women compared to
men during the hyperinsulinaemic euglycaemic clamp (Figure 1). However, the change in
muscle ceramide content during the clamp from baseline did not differ between women
and men (data not shown). There were no differences in muscle glucosylceramide levels
between women and men (basal state: 1.9 [1.0 – 4.9] pmol/mg wet weight vs. 1.4 [1.2
– 3.0] pmol/mg wet weight respectively, P = 0.16 and during the clamp: 1.8 [1.3 – 6.7]
pmol/mg wet weight vs. 2.2 [1.5 – 3.8] pmol/mg wet weight respectively, P = 0.5). In the
basal state, muscle ceramide and glucosylceramide levels did not correlate with plasma
FFA in women and men (ceramide (females): ρ = 0.26; P = 0.47, ceramide (males): ρ =
0.35; P = 0.33 and glucosylceramide (females): ρ = 0.41; P = 0.26, ceramide (males): ρ =
Table 2 REE and oxidation rates for glucose and fat.
basal state hyperinsulinaemic euglycaemic clamp
men (n = 10) women (n = 10) P men (n = 10) women (n = 10) P
REE (kcal/day) 1995 [1840 - 2356] 1511 [884 - 1833] 0.001 1992 [1685 - 2177] 1467 [1150 - 1842] 0.001
Glucose oxidation (μmol/kg•min) 5.0 [0 - 18.3] 3.9 [0 - 7.8] 0.5 20 [8.9 - 33.3] 16.1 [11.7 - 28.3] 0.4
Glucose oxidation (μmol/kg lbm•min) 5.6 [0 - 20.6] 5.0 [0 - 10.6] 0.8 22.2 [11.1 - 37.8] 21.7 [13.9 - 38.3] 0.9
Fat oxidation (μmol/kg•min) 2.0 [0.9 - 2.8] 1.7 [1.3 - 2.1] 0.3 0.7 [0 - 1.5] 0.5 [ 0 - 1.0] 0.4
Fat oxidation (μmol/kg lbm•min) 2.3 [1.0 - 3.0] 2.2 [1.9 - 2.9] 0.9 0.8 [0 - 1.7] 0.6 [0 - 1.3] 0.7
Data are presented as median [minimum - maximum]. REE = resting energy expenditure, lbm = lean body mass.
proefschrift Soeters.indb 35 4-8-2008 9:13:17
36
Table 3 Glucose and lipid metabolism measurements
basal state hyperinsulinaemic euglycaemic clamp
men (n = 10) women (n = 10) P men (n = 10) women (n = 10) P
Glucose (mmol/L) 4.4 [4.0-4.8] 3.9 [2.7-4.5] 0.023 4.9 [4.7-5.2] 5.1 [4.9-5.3] 0.14
EGP (μmol/kg•min) 9.0 [7.3-10.3] 8.0 [ 7.1 - 10.1] 0.07 -* -* -
Rd (μmol/kg•min) 46.8 [41.4-66.5] 48.5 [40.8-72.2] 0.9
FFA (mmol/L) 0.96 [0.72-1.18] 1.26 [0.93-1.54] 0.015 <0.02 <0.02 -
Lipolysis (μmol/kg•min) 4.1 [2.4-5.3] 4.1 [3.6-9.8] 0.3 0.7 [0.1-1.1] 0.9 [0.5-1.5] 0.16
Lipolysis (μmol/kcal) 194 [147-269] 259 [207-526] 0.004 39 [5-61] 50 [29-99] 0.11
Data are presented as median [minimum - maximum]. EGP = endogenous glucose production, Rd = rate of disappearance and FFA = free fatty acids. * During the clamp EGP was completely supressed in both men and women.
Table 4 Glucoregulatory hormones and adiponectin
basal state hyperinsulinaemic euglycaemic clamp
men (n = 10) women (n = 10) P men (n = 10) women (n = 10) P
Insulin (pmol/L) 18 [15-39] 19 [15-29] 0.6 519 [477-624] 551 [433-644] 0.7
Glucagon (ng/L) 75 [44-108] 66 [34-87] 0.6 35 [26-86] 36 [15-51] 0.13
Cortisol (nmol/L) 285 [207-467] 259 [177-357] 0.9 202 [78-311] 207 [95-278] 0.9
Epinephrine (nmol/L) 0.25 [0.11-0.53] 0.17 [0.08-0.30]* 0.11 0.33 [0.15-0.90] 0.17 [0.05-0.26] 0.006
Norepinephrine (nmol/L) 0.68 [0.34-1.52] 0.87 [0.54 - 4.45]* 0.17 1.01 [0.44-3.10] 0.82 [0.56-2.12] 0.4
Adiponectin (μg/ml) 7.7 [3.7-16.0] 16.0 [10.1-21.5] 0.005 6.9 [3.4-15.5] 14.5 [9.3-19.7] 0.005
Data are presented as median [minimum - maximum]. * n = 9.
Figure 1 Muscle ceramide content in men (black boxes) and women (open boxes) in the basal state (men vs. women: P = 1.0) and during the hyperinsulinaemic euglycaemic clamp (men vs. women: P = 0.059).
Figure 2 Protein levels of muscle FAT/CD36 content in men (black boxes) and women (open boxes) in the basal state (men vs. women: P = 0.4) and during the hyperinsulinaemic euglycaemic clamp (men vs. women: P = 0.4).
proefschrift Soeters.indb 36 4-8-2008 9:13:17
Gender differences in fasting
37
Chapter 2
Table 3 Glucose and lipid metabolism measurements
basal state hyperinsulinaemic euglycaemic clamp
men (n = 10) women (n = 10) P men (n = 10) women (n = 10) P
Glucose (mmol/L) 4.4 [4.0-4.8] 3.9 [2.7-4.5] 0.023 4.9 [4.7-5.2] 5.1 [4.9-5.3] 0.14
EGP (μmol/kg•min) 9.0 [7.3-10.3] 8.0 [ 7.1 - 10.1] 0.07 -* -* -
Rd (μmol/kg•min) 46.8 [41.4-66.5] 48.5 [40.8-72.2] 0.9
FFA (mmol/L) 0.96 [0.72-1.18] 1.26 [0.93-1.54] 0.015 <0.02 <0.02 -
Lipolysis (μmol/kg•min) 4.1 [2.4-5.3] 4.1 [3.6-9.8] 0.3 0.7 [0.1-1.1] 0.9 [0.5-1.5] 0.16
Lipolysis (μmol/kcal) 194 [147-269] 259 [207-526] 0.004 39 [5-61] 50 [29-99] 0.11
Data are presented as median [minimum - maximum]. EGP = endogenous glucose production, Rd = rate of disappearance and FFA = free fatty acids. * During the clamp EGP was completely supressed in both men and women.
Table 4 Glucoregulatory hormones and adiponectin
basal state hyperinsulinaemic euglycaemic clamp
men (n = 10) women (n = 10) P men (n = 10) women (n = 10) P
Insulin (pmol/L) 18 [15-39] 19 [15-29] 0.6 519 [477-624] 551 [433-644] 0.7
Glucagon (ng/L) 75 [44-108] 66 [34-87] 0.6 35 [26-86] 36 [15-51] 0.13
Cortisol (nmol/L) 285 [207-467] 259 [177-357] 0.9 202 [78-311] 207 [95-278] 0.9
Epinephrine (nmol/L) 0.25 [0.11-0.53] 0.17 [0.08-0.30]* 0.11 0.33 [0.15-0.90] 0.17 [0.05-0.26] 0.006
Norepinephrine (nmol/L) 0.68 [0.34-1.52] 0.87 [0.54 - 4.45]* 0.17 1.01 [0.44-3.10] 0.82 [0.56-2.12] 0.4
Adiponectin (μg/ml) 7.7 [3.7-16.0] 16.0 [10.1-21.5] 0.005 6.9 [3.4-15.5] 14.5 [9.3-19.7] 0.005
Data are presented as median [minimum - maximum]. * n = 9.
0.53; P = 0.12). No correlation was found between insulin-mediated peripheral glucose
uptake and muscle ceramide or glucosylceramide levels in women and men (ceramide
(females): ρ = 0.14; P = 0.69, ceramide (males): ρ = -0.37; P = 0.29 and glucosylceramide
(females): ρ = -0.05; P = 0.89, glucosylceramide (males): ρ = 0.48; P = 0.16).
FAT/CD36 measurementsThere were no differences in muscle FAT/CD36 content between women and men during
both the basal state and the hyperinsulinaemic euglycaemic clamp (Figure 2). Also, there
was no significant change in muscle FAT/CD36 content between the basal state and the
end of hyperinsulinaemic euglycaemic clamp in both sexes (data not shown).
proefschrift Soeters.indb 37 4-8-2008 9:13:18
38
Discussion
We studied the metabolic adaptation to 38 h of fasting in healthy lean women and men
to elucidate the mechanism behind the well-established finding of lower plasma glucose
levels and higher plasma FFA during fasting in women.
We confirmed earlier reports on lower plasma glucose levels in women during fasting
(1;4;6). The lower basal plasma glucose concentrations in women are probably attributed
to the trend towards a lower EGP. Our findings are not in complete agreement with
previous reports. Two studies showed equal EGP between women and men between
16 to 22 hours of fasting (11;13). In another study, the decline in EGP between 16 and
64 hours of fasting was greater in women than men but absolute EGP did not differ
at both time points between sexes (4). The discrepancy with these studies in finding
a trend towards a lower EGP might be explained by the fact that we studied more
subjects (4;11;13). Moreover, EGP is the major determinant of basal plasma glucose
concentration (38) and the MCR was not different between men and women in the basal
state. This suggests that lower EGP plays a causal role in the lower plasma glucose levels
in women. Notably, gender differences have been described in circulating gluconeogenic
substrates during short-term fasting (1). The difference in EGP between men and women
may be caused by differences in plasma concentrations of ovarian steroids since it was
shown that in female ovariectomized mice, gluconeogenesis was higher compared to
intact control mice (39). To our knowledge, there have been no other reports on gender
differences in EGP after 38 h of fasting.
A remarkable finding in our study was the absence of a difference in peripheral insulin
sensitivity, despite significantly higher plasma FFA in females after fasting for 38 h. So far,
there have been no reports on gender differences regarding insulin sensitivity after short-
term fasting. However, equal or higher insulin sensitivity in women after an overnight
fast has been found (10;15;16;40).
The finding of higher plasma FFA probably results from a higher lipolytic flux. Lipolysis per
kg body weight did not differ between sexes. However, if expressed as flux per REE, women
displayed a higher rate of lipolysis (30;41). The most appropriate way to express lipolysis is
still under discussion (30). The FFA flux can be seen as a function of adipose tissue mass,
but also as a function of fatty acid-consuming tissue requirements. Koutsari et al. argue that
expressing lipolysis per REE is a better alternative, since resting energy requirements are a
major factor determining the rate of FFA release in resting humans (30).
proefschrift Soeters.indb 38 4-8-2008 9:13:18
Gender differences in fasting
39
Chapter 2
Despite higher plasma FFA levels in the basal state, women were equally insulin-
sensitive compared to men, suggesting a relative protection from FFA-induced insulin
resistance. This finding confirms other reports. Perseghin et al. found no differences in
insulin sensitivity between sexes, despite higher levels of plasma as well as intramyocellular
triglycerides in women after an overnight fast (16). Frias et al. demonstrated that women
were less sensitive to FFA-induced insulin resistance during a lipid infusion (12). The
higher plasma FFA levels during short-term fasting in the women we studied, may
result from a decreased FFA-uptake in skeletal muscle, caused by differences in skeletal
muscle fatty acid transporter proteins. The uptake of FFA is thought to depend on,
besides transmembrane concentrations, intracellular fatty acid metabolism, i.e. storage
or oxidation (28;29). We did not find differences in fat oxidation (whole body or per
kilogram lean body mass), which makes increased fatty acid oxidation in skeletal muscle
in our female subjects less likely.
To detect potential differences in FFA-uptake and fatty acid handling we assessed
muscle concentrations of ceramide and glucosylceramide. Ceramide and glucosylceramide
levels in skeletal muscle did not differ between females and males in the basal state,
suggesting that these are not causally involved in the finding of lower fasting plasma
glucose in women. During the clamp, muscle ceramide levels (but not glucosylceramide)
tended to be lower in women compared to men, despite similar Rd. Accordingly; we
found no correlations between muscle ceramide or glucosylceramide and FFA or Rd.
There were no differences between women and men in total muscle FAT/CD36
content in the basal state and during the clamp, which is remarkable since it has been
described earlier that women have higher levels of this lipid-binding protein after an
overnight fast (8) which is in concordance with the findings that the intramyocellular lipid
content during the post absorptive period in women is increased compared to men (16).
Our data suggest that the initial differences in intramyocellular lipids(16) and fatty acid
transporters after an overnight fast (8) are abolished after 38 h of fasting.
Intracellular stored FAT/CD36 is translocated to the plasma membrane, thereby
promoting sarcolemmal long chain fatty acid uptake and this effect is stimulated by
insulin (26). Whether a difference in FAT/CD36 activity (translocation, upregulation)
during fasting exists between males and females is not known, but our data imply such
differences exist. Unfortunately we did not measure the intracellular localization or
activity of the FAT/CD36 protein, which could have clarified higher plasma FFA levels
despite similar total levels of this transporter in skeletal muscle. (25-27;29).
Since we did find higher lipolysis rates in women without an increase in whole body
fat oxidation, the key question remains how the increased plasma FFA in women are
proefschrift Soeters.indb 39 4-8-2008 9:13:18
40
disposed. Uptake of plasma FFA predominantly occurs in liver, adipocytes and muscle.
Shadid et al showed that FFA recycling occurs in the post-absorptive state and that
women display greater efficiency in direct FFA uptake in subcutaneous tissue (femoral
area) (42). The exact role of the adipocyte in FFA uptake under fasting conditions is
currently unknown. The liver may be another site of increased non-oxidative FFA disposal.
Although livers of females do not contain more fat than male livers (43), it is known
that female livers contain more FAT/CD36 and have a greater capacity for FFA uptake
and synthesis of ketone bodies and very low-density lipoproteins-triglycerides (1;14;44).
Despite our data on muscle ceramide and FAT/CD36, we cannot completely rule out
increased FFA disposal in muscle of our female subjects because of earlier mentioned
arguments on possible differences in FAT/CD36 location and activity.
It is unlikely that the difference in plasma epinephrine levels during the clamp can
explain our findings, since the concentration threshold for an effect of epinephrine on
glycaemia has been reported to be between 0.55 and 1.0 nmol/L (45-47).
Finally, women had higher plasma adiponectin levels than men, which may be another
mechanism by which women are relatively protected from FFA-induced insulin resistance.
It has been shown in vitro that adiponectin can stimulate muscle fatty acid oxidation
via stimulation of AMP-kinase, which hypothetically may result in lower intramyocellular
lipid-content with beneficial effects on peripheral insulin sensitivity (48). However, whole
body fat oxidation did not differ between sexes. This challenges the hypothesis of an
important role for adiponectin as protecting factor.
In conclusion, women display lower plasma glucose concentrations and higher plasma
FFA concentrations after 38h of fasting. The former is at least in part explained by lower
endogenous glucose production and the latter by higher rates of lipolysis. Moreover,
women are relatively protected from FFA-induced insulin resistance. This protection does
seem to result neither from differences in fat oxidation rates, muscle concentrations of
ceramide and glucosylceramide, nor differences in total amount of FAT/CD36 in skeletal
muscle. However, a difference in activity of this fatty acid binding protein or intracellular
localization between fasting men and women may result in lower plasma FFA uptake in
women, thereby increasing plasma FFA and decreasing deleterious effects of long chain
fatty acids on insulin signaling.
AcknowledgementsThe authors wish to thank the following colleagues for excellent assistance on laboratory
analyses: A.F.C Ruiter and B.C.E. Voermans (biochemical and stable isotopes analyses),
A. Poppema (sphingo- and glycosphingolipid analyses) and W.A. Coumans (FAT/CD36
proefschrift Soeters.indb 40 4-8-2008 9:13:18
Gender differences in fasting
41
Chapter 2
analyses). J.F.C. Glatz is Netherlands Heart Foundation Professor of Cardiac Metabolism
(grant 2002T049).
Reference List 1. Haymond MW, Karl IE, Clarke WL, Pagliara AS, Santiago JV. Differences in circulating
gluconeogenic substrates during short-term fasting in men, women, and children. Metabolism 1982; 31(1):33-42.
2. Klein S, Sakurai Y, Romijn JA, Carroll RM. Progressive alterations in lipid and glucose metabolism during short-term fasting in young adult men. Am J Physiol Endocrinol Metab 1993; 265(5):E801-E806.
3. Norrelund H, Nielsen S, Christiansen JS, Jorgensen JO, Moller N. Modulation of basal glucose metabolism and insulin sensitivity by growth hormone and free fatty acids during short-term fasting. Eur J Endocrinol 2004; 150(6):779-787.
4. Clore JN, Glickman PS, Helm ST, Nestler JE, Blackard WG. Accelerated decline in hepatic glucose production during fasting in normal women compared with men. Metabolism 1989; 38(11):1103-1107.
5. Hojlund K, Wildner-Christensen M, Eshoj O, Skjaerbaek C, Holst JJ, Koldkjaer O, Moller Jensen D, Beck-Nielsen H. Reference intervals for glucose, {beta}-cell polypeptides, and counterregulatory factors during prolonged fasting. Am J Physiol Endocrinol Metab 2001; 280(1):E50-E58.
6. Merimee TJ, Fineberg SE. Homeostasis during fasting. II. Hormone substrate differences between men and women. J Clin Endocrinol Metab 1973; 37(5):698-702.
7. Binnert C, Koistinen HA, Martin G, Andreelli F, Ebeling P, Koivisto VA, Laville M, Auwerx J, Vidal H. Fatty acid transport protein-1 mRNA expression in skeletal muscle and in adipose tissue in humans. Am J Physiol Endocrinol Metab 2000; 279(5):E1072-E1079.
8. Kiens B, Roepstorff C, Glatz JF, Bonen A, Schjerling P, Knudsen J, Nielsen JN. Lipid-binding proteins and lipoprotein lipase activity in human skeletal muscle: influence of physical activity and gender. J Appl Physiol 2004; 97(4):1209-1218.
9. Blaak E. Gender differences in fat metabolism. Curr Opin Clin Nutr Metab Care 2001; 4(6):499-502.
10. Borissova AM, Tankova T, Kirilov G, Koev D. Gender-dependent effect of ageing on peripheral insulin action. Int J Clin Pract 2005; 59(4):422-426.
11. Corssmit EP, Stouthard JM, Romijn JA, Endert E, Sauerwein HP. Sex differences in the adaptation of glucose metabolism to short-term fasting: effects of oral contraceptives. Metabolism 1994; 43(12):1503-1508.
12. Frias JP, Macaraeg GB, Ofrecio J, Yu JG, Olefsky JM, Kruszynska YT. Decreased susceptibility to fatty acid-induced peripheral tissue insulin resistance in women. Diabetes 2001; 50(6):1344-1350.
13. Mittendorfer B, Horowitz JF, Klein S. Gender differences in lipid and glucose kinetics during short-term fasting. Am J Physiol Endocrinol Metab 2001; 281(6):E1333-E1339.
14. Mittendorfer B. Insulin resistance: sex matters. Curr Opin Clin Nutr Metab Care 2005; 8(4):367-372.
15. Paula FJ, Pimenta WP, Saad MJ, Paccola GM, Piccinato CE, Foss MC. Sex-related differences in peripheral glucose metabolism in normal subjects. Diabete Metab 1990; 16(3):234-239.
16. Perseghin G, Scifo P, Pagliato E, Battezzati A, Benedini S, Soldini L, Testolin G, Del Maschio A, Luzi L. Gender factors affect fatty acids-induced insulin resistance in nonobese humans: effects of oral steroidal contraception. J Clin Endocrinol Metab 2001; 86(7):3188-3196.
17. Boden G. Interaction between free fatty acids and glucose metabolism. Curr Opin Clin Nutr Metab Care 2002; 5(5):545-549.
proefschrift Soeters.indb 41 4-8-2008 9:13:18
42
18. Boden G, Chen X, Ruiz J, White JV, Rossetti L. Mechanisms of fatty acid-induced inhibition of glucose uptake. J Clin Invest 1994; 93(6):2438-2446.
19. Straczkowski M, Kowalska I, Nikolajuk A, Dzienis-Straczkowska S, Kinalska I, Baranowski M, Zendzian-Piotrowska M, Brzezinska Z, Gorski J. Relationship between insulin sensitivity and sphingomyelin signaling pathway in human skeletal muscle. Diabetes 2004; 53(5):1215-1221.
20. Adams JM 2nd, Pratipanawatr T, Berria R, Wang E, DeFronzo RA, Sullards MC, Mandarino LJ. Ceramide content is increased in skeletal muscle from obese insulin-resistant humans. Diabetes 2004; 53(1):25-31.
21. Powell DJ, Turban S, Gray A, Hajduch E, Hundal HS. Intracellular ceramide synthesis and protein kinase Czeta activation play an essential role in palmitate-induced insulin resistance in rat L6 skeletal muscle cells. Biochem J 2004; 382(Pt 2):619-629.
22. Chavez JA, Knotts TA, Wang LP, Li G, Dobrowsky RT, Florant GL, Summers SA. A role for ceramide, but not diacylglycerol, in the antagonism of insulin signal transduction by saturated fatty acids. J Biol Chem 2003; 278(12):10297-10303.
23. Tagami S, Inokuchi Ji J, Kabayama K, Yoshimura H, Kitamura F, Uemura S, Ogawa C, Ishii A, Saito M, Ohtsuka Y, Sakaue S, Igarashi Y. Ganglioside GM3 participates in the pathological conditions of insulin resistance. J Biol Chem 2002; 277(5):3085-3092.
24. Aerts JM, Ottenhoff R, Powlson AS, Grefhorst A, van Eijk M, Dubbelhuis PF, Aten J, Kuipers F, Serlie MJ, Wennekes T, Sethi JK, O’rahilly S, Overkleeft HS. Pharmacological inhibition of glucosylceramide synthase enhances insulin sensitivity. Diabetes 2007;db06-1619.
25. Hamilton JA. Fast flip-flop of cholesterol and fatty acids in membranes: implications for membrane transport proteins. Curr Opin Lipidol 2003; 14(3):263-271.
26. Koonen DPY, Glatz JFC, Bonen A, Luiken JJFP. Long-chain fatty acid uptake and FAT/CD36 translocation in heart and skeletal muscle. Biochimica et Biophysica Acta (BBA) - Molecular and Cell Biology of Lipids 2005; 1736(3):163-180.
27. Schaffer JE. Fatty acid transport: the roads taken. Am J Physiol Endocrinol Metab 2002; 282(2):E239-E246.
28. Mashek DG, Coleman RA. Cellular fatty acid uptake: the contribution of metabolism. Curr Opin Lipidol 2006; 17(3):274-278.
29. Stremmel W, Pohl L, Ring A, Herrmann T. A new concept of cellular uptake and intracellular trafficking of long-chain fatty acids. Lipids 2001; 36(9):981-989.
30. Koutsari C, Jensen MD. Thematic review series: patient-oriented research. Free fatty acid metabolism in human obesity. J Lipid Res 2006; 47(8):1643-1650.
31. Ackermans MT, Pereira Arias AM, Bisschop PH, Endert E, Sauerwein HP, Romijn JA. The quantification of gluconeogenesis in healthy men by (2)H2O and [2-(13)C]glycerol yields different results: rates of gluconeogenesis in healthy men measured with (2)H2O are higher than those measured with [2-(13)C]glycerol. J Clin Endocrinol Metab 2001; 86(5):2220-2226.
32. Ackermans MT, Ruiter AF, Endert E. Determination of glycerol concentrations and glycerol isotopic enrichments in human plasma by gas chromatography/mass spectrometry. Anal Biochem 1998; 258(1):80-86.
33. Groener JEM, Poorthuis BJHM, Kuiper S, Helmond MTJ, Hollak CEM, Aerts JMFG. HPLC for Simultaneous Quantification of Total Ceramide, Glucosylceramide, and Ceramide Trihexoside Concentrations in Plasma. Clin Chem 2007;([Epub ahead of print]):PMID: 17332150 [PubMed - as supplied by publisher].
34. Bonen A, Luiken JJFP, Arumugam Y, Glatz JFC, Tandon NN. Acute Regulation of Fatty Acid Uptake Involves the Cellular Redistribution of Fatty Acid Translocase. J Biol Chem 2000; 275(19):14501-14508.
35. Finegood DT, Bergman RN, Vranic M. Estimation of endogenous glucose production during hyperinsulinemic-euglycemic glucose clamps. Comparison of unlabeled and labeled exogenous glucose infusates. Diabetes 1987; 36(8):914-924.
36. Steele R. Influences of glucose loading and of injected insulin on hepatic glucose output. Ann N Y Acad Sci 1959; 82:420-30.:420-430.
proefschrift Soeters.indb 42 4-8-2008 9:13:18
Gender differences in fasting
43
Chapter 2
37. Frayn KN. Calculation of substrate oxidation rates in vivo from gaseous exchange. J Appl Physiol 1983; 55(2):628-634.
38. Fery F. Role of hepatic glucose production and glucose uptake in the pathogenesis of fasting hyperglycemia in type 2 diabetes: normalization of glucose kinetics by short-term fasting. J Clin Endocrinol Metab 1994; 78(3):536-542.
39. Ahmed-Sorour H, Bailey CJ. Role of ovarian hormones in the long-term control of glucose homeostasis, glycogen formation and gluconeogenesis. Ann Nutr Metab 1981; 25(4):208-212.
40. Yki-Jarvinen H. Sex and insulin sensitivity. Metabolism 1984; 33(11):1011-1015.
41. Nielsen S, Guo Z, Albu JB, Klein S, O’Brien PC, Jensen MD. Energy expenditure, sex, and endogenous fuel availability in humans. J Clin Invest 2003; 111(7):981-988.
42. Shadid S, Koutsari C, Jensen MD. Direct Free Fatty Acid Uptake Into Human Adipocytes In Vivo: Relation to Body Fat Distribution. Diabetes 2007; 56(5):1369-1375.
43. Westerbacka J, Corner A, Tiikkainen M, Tamminen M, Vehkavaara S, Hakkinen AM, Fredriksson J, Yki-Jarvinen H. Women and men have similar amounts of liver and intra-abdominal fat, despite more subcutaneous fat in women: implications for sex differences in markers of cardiovascular risk. Diabetologia 2004; 47(8):1360-1369.
44. Magkos F, Patterson BW, Mohammed BS, Klein S, Mittendorfer B. Women Produce Fewer but Triglyceride-Richer Very Low-Density Lipoproteins than Men. J Clin Endocrinol Metab 2007; 92(4):1311-1318.
45. Chu CA, Sindelar DK, Neal DW, Cherrington AD. Direct effects of catecholamines on hepatic glucose production in conscious dog are due to glycogenolysis. Am J Physiol 1996; 271(1 Pt 1):E127-E137.
46. Clutter WE, Bier DM, Shah SD, Cryer PE. Epinephrine plasma metabolic clearance rates and physiologic thresholds for metabolic and hemodynamic actions in man. J Clin Invest 1980; 66(1):94-101.
47. Cryer PE. Adrenaline: a physiological metabolic regulatory hormone in humans? Int J Obes Relat Metab Disord 1993; 17 Suppl 3:S43-6; discussion S68.:S43-S46.
48. Yoon MJ, Lee GY, Chung JJ, Ahn YH, Hong SH, Kim JB. Adiponectin Increases Fatty Acid Oxidation in Skeletal Muscle Cells by Sequential Activation of AMP-Activated Protein Kinase, p38 Mitogen-Activated Protein Kinase, and Peroxisome Proliferator-Activated Receptor {alpha}. Diabetes 2006; 55(9):2562-2570.
proefschrift Soeters.indb 43 4-8-2008 9:13:18
proefschrift Soeters.indb 44 4-8-2008 9:13:18
Muscle adaptation to short-term fasting in healthy lean humans3Maarten R. Soeters1, Hans P. Sauerwein1, Peter
F. Dubbelhuis2, Johanna E. Groener2, Mariëtte
T. Ackermans3, Eric Fliers1, Johannes M. Aerts2
and Mireille J. Serlie1
1Department of Endocrinology and Metabolism2Department of Medical Biochemistry3Department of Clinical Chemistry, Laboratory
of Endocrinology Academic Medical Center,
University of Amsterdam, Amsterdam, the
Netherlands
J Clin Endocrinol Metab 2008;
93(7):2900-3
proefschrift Soeters.indb 45 4-8-2008 9:13:18
Abstract
Context: It has been demonstrated repeatedly that short-term fasting induces insulin
resistance although the exact mechanism in humans is unknown to date. Intramyocellular
sphingolipids (i.e. ceramide) have been suggested to induce insulin resistance by
interfering with the insulin signaling cascade in obesity.
Objective: To study peripheral insulin sensitivity together with muscle ceramide
concentrations and protein kinase B/AKT phosphorylation after short-term fasting.
Main Outcome Measures and Design: After 14 and 62 hours of fasting, glucose fluxes
were measured before and after a hyperinsulinemic euglycemic clamp. Muscle biopsies
were performed in the basal state and during the clamp to assess muscle ceramide and
protein kinase B/AKT.
Results: Insulin mediated peripheral glucose uptake was significantly lower after 62 h of
fasting compared to 14 h of fasting. Intramuscular ceramide concentrations tended to
increase during fasting. During the clamp the phosphorylation of protein kinase B/AKT
at serine473 in proportion to the total amount of protein kinase B/AKT was significant
lower. Muscle ceramide did not correlate with plasma free fatty acids.
Conclusion: Fasting for 62 h decreases insulin mediated peripheral glucose uptake with
lower phosphorylation of AKT at serine473. AKT may play a regulatory role in fasting
induced insulin resistance. Whether the decrease in AKT can be attributed to the trend
to higher muscle ceramide remains unanswered.
proefschrift Soeters.indb 46 4-8-2008 9:13:18
Muscle adaptation to fasting
47
Chapter 3
Introduction
The incidence of obesity induced insulin resistance and type 2 diabetes mellitus
increases. Lipid infusion studies in healthy subjects showed that increased plasma free
fatty acids (FFA) reduce insulin-mediated glucose uptake (1). Plasma FFA levels correlate
with intramyocellular triglycerides (IMTG) during lipid infusion (2) and IMTG correlate
negatively with peripheral insulin sensitivity (3).
Various metabolites of FFA, such as ceramide, have been suggested to decrease
insulin sensitivity in skeletal muscle (4). Increased muscle ceramide concentrations have
been reported to correlate negatively with insulin sensitivity in obese insulin resistant
subjects (5). Intracellular ceramide synthesis from palmitate is a mechanism by which FFA
decrease insulin-stimulated phosphorylation of protein kinase B/AKT (AKT) (6).
During short-term fasting, plasma FFA (7) as well as intramyocellular lipid stores
increase (8). Also, fasting induces insulin resistance although the exact mechanism is still
unknown (9;10). Fasting increases intramyocellular ceramide levels in rats (11), but this
has not been investigated in lean healthy humans.
Therefore, we studied peripheral glucose metabolism during hyperinsulinemic
euglycemic clamp conditions in healthy lean subjects after 14 and 62 h of fasting in
relation to muscle ceramide and phosphorylation of AKT at serine473 (pAKT-ser473). We
hypothesized that fasting increased muscle ceramide and decreased peripheral insulin
sensitivity via lower muscle pAKT-ser473 levels.
Subjects and methods
SubjectsAfter informed consent, eight healthy, non smoking, male volunteers were included. The
study was approved by the Medical Ethical Committee of the Academic Medical Center.
ProtocolSubjects were studied after 14 and 62 h of fasting, separated by at least a week. Subjects
fasted from 2000 h the evening before the first study day and from 2000 h three days
before the second study day. Glucose kinetics (tracer: [6,6-2H2]glucose; >99% enriched;
Cambridge Isotopes, Andover, USA: prime, 8.8 μmol/kg; continuous, 0.11 μmol/kg·min),
FFA and glucoregulatory hormones were measured in the basal state and after a 5 h
hyperinsulinemic euglycemic clamp (insulin infusion: 60mU/m2·min; Actrapid 100 IU/
proefschrift Soeters.indb 47 4-8-2008 9:13:18
48
ml; Novo Nordisk Farma B.V., Alphen aan den Rijn, the Netherlands) as described earlier
(12). Insulin infusion rates were chosen to completely suppress endogenous glucose
production (13).
Indirect calorimetry (O2 consumption and CO2 production) and muscle biopsies were
performed at the end of both basal state and clamp as described previously (12).
Analytical ProceduresPlasma glucose and FFA concentrations were measured as reported earlier (12). [6,6-2H2]
glucose enrichment was measured as described previously (12).
Insulin, cortisol, glucagon and catecholamines were determined as described previously
(12). Soluble tumor necrosis factor receptors (sTNF-R) I and II were determined with an
EASIA kit (Biosource Europe S.A., Belgium).
Ceramide in muscle biopsies was measured as described previously (12).
Muscle immunoblots were visualized by enhanced chemiluminescence (ECL). Chemicals
for ECL were from Sigma (St. Louis, MO, USA). Phosphospecific anti-AKT-ser473,
phosphospecific anti-glycogen synthase kinase-3-ser9 (GSK), total anti-AKT and total anti-
eIF4E (loading control) were from Cell Signaling (Boston, MA, USA). Phosphospecific
anti-AS160-thr642 was from GeneTex Inc. (San Antonio, TX, USA). Twenty mg of muscle
tissue was taken up in 300μl of ice-cold lysis buffer (20 mM Tris (pH 7.5), 50 mM NaCl, 250
mM sucrose, 50mM NaF, 5mM Na4P2O7, 1mM DTT, 1,0% Triton X-100) supplemented
with cocktail protease inhibitor tablets. Cell lysate was cleared by centrifugation for 15
min at 4°C. Cell protein was determined and separated by SDS-PAGE. A standard Western
blotting procedure was performed and polyvinylidene fluoride blots were incubated with
appropriate antibodies. Results are presented as fold increase compared to control after
14h fasting.
Calculations and statisticsEndogenous glucose production (EGP) and peripheral glucose uptake (rate of
disappearance/Rd) were calculated with modified forms of the Steele Equations as
described (12). Glucose oxidation was calculated as reported previously (12).
Comparisons and correlations were performed with the Wilcoxon Signed Rank test and
Spearman’s rank correlation analysis (ρ) respectively. The statistical software program
version 12.0.1 (SPSS Inc, Chicago, IL) was used for statistical analysis. Data are presented
as median [minimum -maximum].
proefschrift Soeters.indb 48 4-8-2008 9:13:18
Muscle adaptation to fasting
49
Chapter 3
Results
Anthropometric characteristicsSubject characteristics were: age: 23 (20 - 26) yrs; weight 70.1 (62.5 – 75.5) kg after 14
h and 69.0 (60.0 – 72.8) kg after 62 h of fasting, P = 0.012; BMI 20.9 (19.2 – 23.3) kg/
m2 after 14 h and 20.3 (18.3 – 22.6) kg/m2 after 62 h of fasting, P = 0.011.
Glucose kinetics, FFA and glucoregulatory hormones (Table 1)Basal plasma glucose concentrations, glucose oxidation and EGP were significantly lower
after 62 h of fasting, whereas basal plasma FFA increased significantly.
No differences were found in plasma glucose concentrations between clamps. Rd
was significantly lower after 62 h of fasting. Non-oxidative (NOGD) glucose disposal and
oxidative glucose disposal during the clamp was lower after 62 h. Glucose oxidation and
NOGD expressed as percentage of Rd did not differ between clamps. Plasma FFA were
equally suppressed during both clamps.
Insulin levels were lower after 62 h of fasting in the basal state and during the clamp.
Glucagon levels were higher in the basal state and tended to be higher during the clamp
after 62 h of fasting. Fasting did not influence plasma cortisol and norepinephrine levels.
Plasma epinephrine concentrations, however, were higher after 62 h of fasting in the
basal state but not during the clamp. sTNF-RI and II did not change.
Muscle measurementsMuscle ceramide concentrations in the basal state tended to be higher after 62 h of
fasting: 38.0 [25.2 - 54.9] pmol/mg wet weight vs. 29.5 [14.0 - 58.9] pmol/mg wet
weight respectively (P = 0.069). During the clamp no differences were found between
62 h and 14 h of fasting: 35.3 [26.2 - 84.6] pmol/mg wet weight vs. 32.2 [25.9 – 54.3]
pmol/mg wet weight respectively, P = 0.5.
Insulin-mediated peripheral glucose uptake and muscle ceramide levels did not
correlate after 62 h of fasting: ρ = -0.12; P = 0.78. Muscle ceramide and plasma FFA
levels showed no correlation (ρ = -0.30; P = 0.47).
pAKT-ser473 increased significantly during both clamps (Figure 1). A significant lower
ratio of pAKT-ser473 to total AKT (pAKT-ser473/tAKT) was observed during the clamp
after 62 h of fasting vs. 14 h, but not in the basal state. The increase of pAKT-ser473/tAKT
was lower after 62 h of fasting.
pAS160-thr642 increased significantly during both clamps, but lower pAS160-thr642
was observed during the clamp after 62 h of fasting.
proefschrift Soeters.indb 49 4-8-2008 9:13:18
50
Figure 1, panel A pAKT-Ser473 during the basal state (P = 0.3) and during the hyperinsulinemic euglycemic clamp (**P = 0.069) after 14 h (open box plots) and 62h (grey box blots) of fasting. *P = 0.012 and 0.017 for the increase in pAKT-Ser473 during the clamps after 14 and 62 h of fasting respectively.Figure 1, panel B pAKT-Ser473 in proportion to total AKT (pAKT-ser473/tAKT) (n = 6) during the basal state (P = 0.3) and during the hyperinsulinemic euglycemic clamp (***P = 0.028) after 14 h (open box plots) and 62h (grey box blots) of fasting. *P = 0.028 and 0.028 for the increase in pAKT-ser473/tAKT during the clamps after 14 and 62 h of fasting respectively.Figure 1, panel C pGSK-3-ser9 (n = 6) during the basal state (P = 0.3) and during the hyperinsulinemic euglycemic clamp (P = 0.3) after 14 h (open box plots) and 62h (grey box blots) of fasting. *P = 0.028 and 0.028 for the increase in pGSK-3-ser9 during the clamps after 14 and 62 h of fasting respectively.Figure 1, panel D pAS160-Thr642 during the basal state (P = 0.7) and during the hyperinsulinemic euglycemic clamp (***P = 0.017) after 14 h (open box plots) and 62h (grey box blots) of fasting. *P = 0.012 and 0.012 for the increase in pAS160-Thr642 during the clamps after 14 and 62 h of fasting respectively.
proefschrift Soeters.indb 50 4-8-2008 9:13:18
Muscle adaptation to fasting
51
Chapter 3
pGSK-3-ser9 was not different between basal states or clamps, but increased
significantly during the clamps.
Discussion
We studied the adaptation to 62 h of fasting in healthy lean men to explore the mechanism
underlying the fasting induced decrease in peripheral insulin sensitivity.
Our study confirms reports on lower glucose concentrations, EGP and peripheral insulin
sensitivity after fasting (9;10). The lower NOGD after fasting, supports data by Bergman
et al (9). However not all studies detected changes in NOGD during fasting (14). If
glucose oxidation and NOGD were expressed as percentage of Rd, no differences were
found, suggesting that the intracellular fate of glucose remains intact despite decreased
peripheral glucose uptake.
To further explore the fasting induced insulin resistance, we examined muscle ceramide.
The trend towards increased muscle ceramide levels in the basal state after 62h of fasting
suggests a fasting effect, but hinders a definite assumption. Muscle ceramide is mainly
derived from de novo synthesis from serine and palmitate (6). We found no correlation
of muscle ceramide with plasma FFA levels, suggesting that de novo synthesis is not the
denominator of muscle ceramide during short-term fasting (12).
Sphingomyelin hydrolysis, another pathway resulting in ceramide generation, occurs
under stress stimuli like TNFα (6). However, we found no changes in plasma sTNF-RI
and II. Also, we earlier reported no changes in inflammatory parameters during fasting
(10). Therefore, the trend for higher muscle ceramide levels within skeletal muscle after
fasting remains unexplained.
AKT is a 56 kD serine/threonine kinase and a mediator of many insulin effects and its
regulation is complex (15). To be activated, AKT is translocated to the plasma membrane
via its PH domain that binds PI3,4,5P3, the product of phosphoinositol-3-kinase (PI3K).
Here phosphorylation of serine473 by 3-phosphoinositide dependent kinase (PDK) 2 (16)
and threonine308 (thr308) by PDK1 occur (15). We found a significant lower ratio pAKT-
ser473/tAKT after 62 h of fasting during the clamp but not in the basal state. Bergman
et al, reported no difference in pAKT-ser473 and the ratio pAKT-ser473/tAKT between 12
and 48 h of fasting. This may be explained by differences in fasting duration. Remarkably,
Bergman et al showed no increase of pAKT-ser473 during hyperinsulinemia as reported
earlier (17;18). Intriguingly, lipid infusion in healthy men induces peripheral insulin
resistance without effect on pAKT-ser473 (17). Another study found no differences in
proefschrift Soeters.indb 51 4-8-2008 9:13:18
52
Table 1 Glucose kinetics, FFA and glucoregulatory hormones
basal state hyperinsulinaemic euglycaemic clamp
14h(n = 8)
62h(n = 8)
P 14h(n = 8)
62h(n = 8)
P
Glucose oxidation (μmol/kg•min) 6.8 [3.8 - 14.1] 0.8 [0 - 2.2] 0.012 21.1 [18.1 - 23.3] 13.5 [6.9 - 17.4] 0.012
Glucose (mmol/liter) 5.0 [4.5 - 5.6] 3.7 [3.3 - 4.1] 0.011 5.1 [4.9 - 5.3] 4.9 [4.5 - 5.1] 0.107
EGP (μmol/kg•min) 11.8 [8.9 - 19.7] 8.3 [7.0 - 8.7] 0.012 -* -* -
Rd (μmol/kg•min) - - - 60.5 [45.1 - 79.3] 44.4 [37.5 - 51.7] 0.018
NOGD (μmol/kg•min) - - - 39.7 [23.1 - 56.0] 31.6 [20.0 - 39.6] 0.050
Glucose oxidation (% of Rd) - - - 32.4 [26.4 - 49.7] 30.0 [17.0 - 46.4] 0.16
NOGD (% of Rd) - - - 67.6 [50.3 - 73.6] 70.0 [83.0 - 53.6] 0.16
FFA (mmol/liter) 0.33 [0.19 - 0.95] 1.1 [0.85 - 1.24] 0.015 <0.02 <0.02 -
Insulin (pmol/liter) 30 [18 - 58] 15 [15 - 20] 0.012 642 [416 - 715] 533 [383 - 663] 0.012
Glucagon (ng/liter) 57 [40 - 72] 105 [65 - 148] 0.012 33 [18 - 39] 37 [25 - 60] 0.069
Cortisol (nmol/liter) 291 [223 - 354] 274 [221 - 497] 0.263 151 [112 - 245] 190 [147 - 281] 0.161
Epinephrine (nmol/liter) 0.10 [0.05 - 0.23] 0.18 [0.07 - 0.40] 0.063 0.13 [0.05 - 0.19] 0.14 [0.05 - 0.23] 0.734
Norepinephrine (nmol/liter) 0.52 [0.20 - 0.84] 0.63 [0.20 - 0.92] 0.237 0.60 [0.32 - 0.92] 0.62 [0.18 - 0.91] 0.161
sTNF-RI (ng/ml) 1.2 [1.0 - 1.5] 1.2 [0.9 - 1.3] 0.666 1.3 [1.2 - 1.5] 1.3 [1.0 - 1.5] 0.862
sTNF-RII (ng/ml) 3.6 [2.8 - 4.5] 3.8 [2.7 - 4.9] 0.344 3.6 [2.5 - 5.2] 3.6 [2.7 - 4.6] 0.482
Data are presented as median [minimum - maximum]. EGP, endogenous glucose production; NOGD, non-oxidative glucose disposal; FFA, free fatty acids. * During the clamps, EGP and FFA were completely suppressed. NOGD = non-oxidative glucose disposal during the clamp.
pAKT-ser473 between patients with type 2 diabetes and healthy matched controls (18).
This indicates that the impairment in insulin signaling during fasting differs from the
impairment during elevation of plasma FFA by lipid infusion or obesity. Whether the lower
pAKT-ser473/tAKT during hyperinsulinemia is attributed to the trend to higher muscle
ceramide levels in the basal state remains unanswered, since we found no correlation
of ceramide levels with peripheral insulin sensitivity in line with previous observations
(12;19). Other lipid mediators such as diacylglycerol (DAG) or GM3 may interfere with
the insulin signaling cascade (4): fasting increases muscle DAG in animals (11).
AKT phosphorylates both AS160 and GSK. AS160 is involved in the insulin induced
translocation of GLUT4; moreover insulin mediated phosphorylation of AS160 was
shown to be decreased in patients with type 2 diabetes (18). Phosphorylation of GSK
by AKT stimulates glycogen synthesis (20). The equal pGSK-3-ser9 during the clamps
are in line with the NOGD (as percentage of Rd) and suggests differential regulation of
downstream events by AKT since pAS160-thr642 and peripheral insulin sensitivity were
lower after 62 h of fasting.
proefschrift Soeters.indb 52 4-8-2008 9:13:19
Muscle adaptation to fasting
53
Chapter 3
Table 1 Glucose kinetics, FFA and glucoregulatory hormones
basal state hyperinsulinaemic euglycaemic clamp
14h(n = 8)
62h(n = 8)
P 14h(n = 8)
62h(n = 8)
P
Glucose oxidation (μmol/kg•min) 6.8 [3.8 - 14.1] 0.8 [0 - 2.2] 0.012 21.1 [18.1 - 23.3] 13.5 [6.9 - 17.4] 0.012
Glucose (mmol/liter) 5.0 [4.5 - 5.6] 3.7 [3.3 - 4.1] 0.011 5.1 [4.9 - 5.3] 4.9 [4.5 - 5.1] 0.107
EGP (μmol/kg•min) 11.8 [8.9 - 19.7] 8.3 [7.0 - 8.7] 0.012 -* -* -
Rd (μmol/kg•min) - - - 60.5 [45.1 - 79.3] 44.4 [37.5 - 51.7] 0.018
NOGD (μmol/kg•min) - - - 39.7 [23.1 - 56.0] 31.6 [20.0 - 39.6] 0.050
Glucose oxidation (% of Rd) - - - 32.4 [26.4 - 49.7] 30.0 [17.0 - 46.4] 0.16
NOGD (% of Rd) - - - 67.6 [50.3 - 73.6] 70.0 [83.0 - 53.6] 0.16
FFA (mmol/liter) 0.33 [0.19 - 0.95] 1.1 [0.85 - 1.24] 0.015 <0.02 <0.02 -
Insulin (pmol/liter) 30 [18 - 58] 15 [15 - 20] 0.012 642 [416 - 715] 533 [383 - 663] 0.012
Glucagon (ng/liter) 57 [40 - 72] 105 [65 - 148] 0.012 33 [18 - 39] 37 [25 - 60] 0.069
Cortisol (nmol/liter) 291 [223 - 354] 274 [221 - 497] 0.263 151 [112 - 245] 190 [147 - 281] 0.161
Epinephrine (nmol/liter) 0.10 [0.05 - 0.23] 0.18 [0.07 - 0.40] 0.063 0.13 [0.05 - 0.19] 0.14 [0.05 - 0.23] 0.734
Norepinephrine (nmol/liter) 0.52 [0.20 - 0.84] 0.63 [0.20 - 0.92] 0.237 0.60 [0.32 - 0.92] 0.62 [0.18 - 0.91] 0.161
sTNF-RI (ng/ml) 1.2 [1.0 - 1.5] 1.2 [0.9 - 1.3] 0.666 1.3 [1.2 - 1.5] 1.3 [1.0 - 1.5] 0.862
sTNF-RII (ng/ml) 3.6 [2.8 - 4.5] 3.8 [2.7 - 4.9] 0.344 3.6 [2.5 - 5.2] 3.6 [2.7 - 4.6] 0.482
Data are presented as median [minimum - maximum]. EGP, endogenous glucose production; NOGD, non-oxidative glucose disposal; FFA, free fatty acids. * During the clamps, EGP and FFA were completely suppressed. NOGD = non-oxidative glucose disposal during the clamp.
A remarkable finding in our study was the lower insulin levels during the clamp after
62 h of fasting. Insulin infusions were almost identical during both clamps. It is unlikely
that endogenous insulin secretion was stimulated at these euglycemic conditions.
Plasma clearance of infused insulin is mainly renal in contrast to the first pass effect
of endogenous insulin by the liver (21). Earlier studies showed equal insulin levels and
lower Rd after fasting, making fasting induced insulin resistance widely accepted (9). The
insulin dose response curve negates, that the different plasma insulin levels account for
differences in Rd (13).
In conclusion, short-term fasting induces peripheral insulin resistance of glucose uptake
while the muscle fate of glucose stays intact. Muscle ceramide tends to increase during
fasting. The decreased peripheral glucose uptake is explained by a decrease in pAKT-
ser473/tAKT and pAS160-thr642 during the clamp.
Since studies in obesity induced insulin resistance and type 2 diabetes mellitus have
not shown effects on pAKT-ser473, it is possible that pAKT-ser473 is involved in the
physiological adaptation to fasting, inducing a reduction in peripheral glucose uptake
and protecting the body from hypoglycemia.
proefschrift Soeters.indb 53 4-8-2008 9:13:19
54
AcknowledgementsWe thank A.F.C Ruiter and A. Poppema for excellent assistance on laboratory analyses.
Reference List 1. Belfort R, Mandarino L, Kashyap S, Wirfel K, Pratipanawatr T, Berria R, DeFronzo RA, Cusi K.
Dose-Response Effect of Elevated Plasma Free Fatty Acid on Insulin Signaling. Diabetes 2005; 54(6):1640-1648.
2. Boden G, Lebed B, Schatz M, Homko C, Lemieux S. Effects of Acute Changes of Plasma Free Fatty Acids on Intramyocellular Fat Content and Insulin Resistance in Healthy Subjects. Diabetes 2001; 50(7):1612-1617.
3. Krssak M, Falk Petersen K, Dresner A, DiPietro L, Vogel SM, Rothman DL, Roden M, Shulman GI. Intramyocellular lipid concentrations are correlated with insulin sensitivity in humans: a 1H NMR spectroscopy study. Diabetologia 1999; 42(1):113-116.
4. Holland WL, Knotts TA, Chavez JA, Wang LP, Hoehn KL, Summers SA. Lipid Mediators of Insulin Resistance. Nutrition Reviews 2007; 65:39-46.
5. Straczkowski M, Kowalska I, Nikolajuk A, Dzienis-Straczkowska S, Kinalska I, Baranowski M, Zendzian-Piotrowska M, Brzezinska Z, Gorski J. Relationship between insulin sensitivity and sphingomyelin signaling pathway in human skeletal muscle. Diabetes 2004; 53(5):1215-1221.
6. Summers SA. Ceramides in insulin resistance and lipotoxicity. Prog Lipid Res 2006; 45(1):42-72.
7. Klein S, Sakurai Y, Romijn JA, Carroll RM. Progressive alterations in lipid and glucose metabolism during short-term fasting in young adult men. Am J Physiol Endocrinol Metab 1993; 265(5):E801-E806.
8. Stannard SR, Thompson MW, Fairbairn K, Huard B, Sachinwalla T, Thompson CH. Fasting for 72 h increases intramyocellular lipid content in nondiabetic, physically fit men. Am J Physiol Endocrinol Metab 2002; 283(6):E1185-E1191.
9. Bergman BC, Cornier MA, Horton TJ, Bessesen DH. Effects of fasting on insulin action and glucose kinetics in lean and obese men and women. Am J Physiol Endocrinol Metab 2007; 293(4):E1103-E1111.
10. van der Crabben SN, Allick G, Ackermans MT, Endert E, Romijn JA, Sauerwein HP. Prolonged Fasting Induces Peripheral Insulin Resistance, Which Is Not Ameliorated by High-Dose Salicylate. J Clin Endocrinol Metab 2008; 93(2):638-641.
11. Turinsky J, Bayly BP, O’Sullivan DM. 1,2-Diacylglycerol and ceramide levels in rat liver and skeletal muscle in vivo. Am J Physiol 1991; 261(5 Pt 1):E620-E627.
12. Soeters MR, Sauerwein HP, Groener JE, Aerts,JM, Ackermans MT, Glatz JF, Fliers E, Serlie MJ. Gender-Related Differences in the Metabolic Response to Fasting. J Clin Endocrinol Metab 2007; 92(9):3646-3652.
13. Rizza RA, Mandarino LJ, Gerich JE. Dose-response characteristics for effects of insulin on production and utilization of glucose in man. Am J Physiol Endocrinol Metab 1981; 240(6):E630-E639.
14. Webber J, Taylor J, Greathead H, Dawson J, Buttery PJ, Macdonald IA. Effects of fasting on fatty acid kinetics and on the cardiovascular, thermogenic and metabolic responses to the glucose clamp. Clin Sci (Lond) 1994; 87(6):697-706.
15. Scheid MP, Woodgett JR. Unravelling the activation mechanisms of protein kinase B/Akt. FEBS Letters 2003; 546(1):108-112.
16. Dong LQ, Liu F. PDK2: the missing piece in the receptor tyrosine kinase signaling pathway puzzle. Am J Physiol Endocrinol Metab 2005; 289(2):E187-E196.
proefschrift Soeters.indb 54 4-8-2008 9:13:19
Muscle adaptation to fasting
55
Chapter 3
17. Tsintzas K, Chokkalingam K, Jewell K, Norton L, Macdonald IA, Constantin-Teodosiu D. Elevated Free Fatty Acids Attenuate the Insulin-Induced Suppression of PDK4 Gene Expression in Human Skeletal Muscle: Potential Role of Intramuscular Long-Chain Acyl-Coenzyme A. J Clin Endocrinol Metab 2007; 92(10):3967-3972.
18. Karlsson HKR, Zierath JR, Kane S, Krook A, Lienhard GE, Wallberg-Henriksson H. Insulin-Stimulated Phosphorylation of the Akt Substrate AS160 Is Impaired in Skeletal Muscle of Type 2 Diabetic Subjects. Diabetes 2005; 54(6):1692-1697.
19. Serlie MJ, Meijer AJ, Groener JE, Duran M, Endert E, Fliers E, Aerts JM, Sauerwein HP. Short-Term Manipulation of Plasma Free Fatty Acids Does Not Change Skeletal Muscle Concentrations of Ceramide and Glucosylceramide in Lean and Overweight Subjects. J Clin Endocrinol Metab 2007; 92(4):1524-1529.
20. Doble BW, Woodgett JR. GSK-3: tricks of the trade for a multi-tasking kinase. J Cell Sci 2003; 116(7):1175-1186.
21. Duckworth WC, Bennett RG, Hamel FG. Insulin Degradation: Progress and Potential. Endocr Rev 1998; 19(5):608-624.
proefschrift Soeters.indb 55 4-8-2008 9:13:19
proefschrift Soeters.indb 56 4-8-2008 9:13:19
Decreased suppression of muscle fatty acid oxidation by hyperinsulinemia during fasting4Maarten R. Soeters1, Hans P. Sauerwein1,
Marinus Duran2, Ronald J. Wanders2, Mariëtte
T. Ackermans3, Eric Fliers1, Sander M. Houten2
and Mireille J. Serlie1
1Department of Endocrinology and Metabolism2 Department of Clinical Chemistry, Laboratory
Genetic Metabolic Diseases3 Department of Clinical Chemistry, Laboratory
of Endocrinology Academic Medical Center,
University of Amsterdam, Amsterdam, the
Netherlands
Submitted
proefschrift Soeters.indb 57 4-8-2008 9:13:19
Abstract
Context: The transition from the fed to the fasted resting state is characterized by, amongst
others, changes in lipid metabolism and peripheral insulin resistance. Acylcarnitines have
been suggested to play a role in insulin resistance besides other long-chain fatty acid
metabolites. Plasma levels of long-chain acylcarnitines increase during fasting, but this is
unknown for muscle long-chain acylcarnitines.
Objective: We studied whether muscle long-chain acylcarnitines increase during fasting
and their relation with glucose/fat oxidation and insulin sensitivity in lean healthy
humans.
Main Outcome Measures and Design: After 14 and 62 hours of fasting, glucose fluxes,
substrate oxidation, plasma and muscle acylcarnitines were measured before and during
a hyperinsulinemic euglycemic clamp.
Results: Hyperinsulinemia decreased long-chain muscle acylcarnitines after 14 hours of
fasting but not after 62 hours of fasting. In both the basal state and during the clamp
glucose oxidation was lower and fatty acid oxidation was higher after 62 hours vs. 14 hours
of fasting. Absolute changes in glucose and fat oxidation in the basal vs. hyperinsulinemic
state were not different. Muscle long-chain acylcarnitines did not correlate with glucose
oxidation, fatty acid oxidation or insulin-mediated peripheral glucose uptake.
Conclusion: After 62 hours of fasting, the suppression of muscle long-chain acylcarnitines
by insulin was attenuated compared to 14 hours of fasting. Muscle long-chain acylcarnitines
do not unconditionally reflect fatty acid oxidation. The higher fatty acid oxidation during
hyperinsulinemia after 62 vs. 14 hours of fasting although the absolute decrease in FAO
was not different, suggests a different insulin-regulated set point.
proefschrift Soeters.indb 58 4-8-2008 9:13:19
Muscle fatty acid oxidation in fasting
59
Chapter 4
Introduction
Short-term fasting can be defined as the first 72 hours of starvation in which progressive
alterations in lipid and glucose metabolism occur (1;2). The adaptation to short-term
fasting is characterized by, amongst others, an increase in lipolysis with concomitant
increases in plasma free fatty acids (FFA) and fatty acid oxidation (FAO) (1) and a decrease
in peripheral insulin sensitivity and carbohydrate oxidation (CHO)(2-4).
To be oxidized, activated long-chain fatty acids can only cross the mitochondrial
membranes as acylcarnitines (ACs)(5). The coupling of an activated long-chain fatty acid
to carnitine (3-hydroxy-4-N,N,N-trimethylaminobutyric acid) is catalyzed by carnitine-
acyl(palmitoyl)transferase 1 (CPT1) on the outer mitochondrial leaflet (6). CPT1 is considered
to be the rate-limiting enzyme for long-chain fatty acid entry into the mitochondria and
subsequent oxidation (5;7). Furthermore CPT1 activity increases during fasting in animal
studies (8). Inside the mitochondrion, the AC is activated to acyl-CoA again by CPT2. The
released carnitine is exchanged for a new incoming AC by the mitochondrial membrane
protein carnitine-acylcarnitine-translocase (CACT) (6).
Plasma ACs are thought to reflect the mitochondrial acyl-CoA pool and AC profile
analysis is the current standard for the diagnosis of FAO disorders at the metabolite level
(9;10). Recently, muscle ACs have been implicated in insulin resistance via a currently
unknown mechanism (8;11;12). These studies suggested that increased beta-oxidation
outpaces the tricarboxylic acid cycle (TCA) with subsequent inhibition of complete fatty
acid oxidation via a high energy redox state (rising NADH/NAD+ and acetyl-CoA/CoA
ratios). The accumulation of metabolic by-products (e.g. acylcarnitines) would then
activate stress kinases or other signals, interfering with insulin action (11). However, how
ACs directly affect insulin mediated glucose uptake is currently unraveled.
Fasting increases plasma long-chain ACs (13-15), but it is unknown, whether muscle
long-chain ACs increase during short-term fasting in humans. Animal studies showed
increased muscle long-chain AC levels during fasting (16;17). Such an increase of ACs
during fasting would match increased lipid oxidation and decreased peripheral insulin
sensitivity (1;3;4).
In this study, we examined the association of muscle ACs during short-term fasting with
glucose and fat metabolism. Healthy lean subjects were studied before and after short-
term fasting in both the basal state and during a hyperinsulinemic euglycemic clamp. We
hypothesized that fasting induced an increase of whole body FAO resulting in an increase
in muscle ACs which would explain the expected peripheral insulin resistance.
proefschrift Soeters.indb 59 4-8-2008 9:13:19
60
Subjects and methods
SubjectsWe studied healthy lean male volunteers who participated in a study on fasting induced
insulin resistance (4). Subjects were in self-reported good health, confirmed by medical
history and physical examination. Criteria for inclusion were 1) absence of a family
history of diabetes; 2) age 18–35 yr; 3) Caucasian race; 4) BMI 20-25 kg/m2; 5) normal
oral glucose tolerance test according to the ADA criteria (18); 6) normal routine blood
examination; 7) no excessive sport activities, i.e. < 3 times per week; and 8) no medication.
Written informed consent was obtained from all subjects after explanation of purposes,
nature, and potential risks of the study. The study was approved by the Medical Ethical
Committee of the Academic Medical Center of the University of Amsterdam.
Experimental protocolSubjects were studied twice: after 14 and 62 h of fasting. Study days were separated by
at least a week. Subjects were fasting from 2000 h the evening before the first study day
and from 2000 h three days before the second study day until the end of the study days.
They were allowed to drink water only.
After admission to the metabolic unit at 0730 h, a catheter was inserted into an
antecubital vein for infusion of stable isotope tracers, insulin and glucose. Another
catheter was inserted retrogradely into a contralateral hand vein and kept in a thermo-
regulated (60ºC) plexiglas box for sampling of arterialized venous blood. Saline was
infused as NaCl 0.9% at a rate of 50 mL/h to keep the catheters patent. [6,6-2H2]glucose
and [1,1,2,3,3-2H5]glycerol were used as tracers (>99% enriched; Cambridge Isotopes,
Andover, USA) to study glucose kinetics and lipolysis (total triglyceride hydrolysis)
respectively.
At T = 0 h (0800 h), blood samples were drawn for determination of background
enrichments and a primed continuous infusion of both isotopes was started: [6,6-2H2]
glucose (prime, 8.8 μmol/kg; continuous, 0.11 μmol/kg·min) and [1,1,2,3,3-2H5]glycerol
(prime, 1.6 μmol/kg; continuous, 0.11 μmol/kg·min) and continued until the end of the
study. After an equilibration period of two hours (14 h of fasting), 3 blood samples were
drawn for glucose and glycerol enrichments and 1 for glucoregulatory hormones, FFA
and plasma AC levels. Thereafter (T = 3 h), infusions of insulin (60mU/m2·min) (Actrapid
100 IU/ml; Novo Nordisk Farma B.V., Alphen aan den Rijn, the Netherlands) and glucose
20% (to maintain a plasma glucose level of 5 mmol/L) were started. [6,6-2H2]glucose was
added to the 20% glucose solution to achieve glucose enrichments of 1% to approximate
proefschrift Soeters.indb 60 4-8-2008 9:13:19
Muscle fatty acid oxidation in fasting
61
Chapter 4
the values for enrichment reached in plasma and thereby minimizing changes in isotopic
enrichment due to changes in the infusion rate of exogenous glucose (19). Plasma
glucose levels were measured every 5 min at the bedside. At T = 8 h, 5 blood samples
were drawn at 5-min intervals for determination of glucose and glycerol enrichments.
Another blood sample was drawn for determination of glucoregulatory hormones, FFA
and plasma AC levels.
Subjects were studied under the same conditions after 62h of fasting. Volunteers
were allowed to drink water ad libitum. To prevent sodium and potassium depletion
during fasting, subjects were supplied with oral sodium chloride 80mmol per day (Tablets,
In-House Pharmacist, AMC, Amsterdam, the Netherlands) and oral potassium chloride
40mmol per day (Slow-K, Novartis BV, Arnhem, the Netherlands).
Indirect calorimetry and muscle biopsiesOxygen consumption (VO2) and CO2 production (VCO2) were measured continuously
during the final 20 min of both the basal state and the clamp by indirect calorimetry using
a ventilated hood system (Sensormedics model 2900; Sensormedics, Anaheim, CA).
Muscle biopsies were performed to assess muscle AC concentrations at the end of both
the basal state and the clamp. Biopsies was performed under local anaesthesia (lidocaine
20 mg/ml; Fresenius, Kabi, Den Bosch, The Netherlands) using a Pro-Mag I biopsy needle
(MDTECH, Gainesville, FL). Biopsy specimens were quickly washed in a buffer (0.9%
NaCl/28.3g/liter HEPES) to remove blood, inspected for fat or fascia content, dried on
gauze swabs, and subsequently stored in liquid nitrogen until analysis.
Glucose and lipid metabolism measurementsPlasma glucose and FFA concentrations were measured as described earlier (4). [6,6-2H2]
glucose enrichment was measured as described earlier (20). [6,6-2H2]glucose enrichment
(tracer/tracee ratio) intra-assay variation: 0.5-1%; inter-assay variation 1%; detection
limit: 0.04%. [1,1,2,3,3-2H5]glycerol enrichment was determined as described earlier
(21). Intra-assay variation glycerol: 1-3%, [1,1,2,3,3-2H5]glycerol: 4%; inter-assay variation
glycerol: 2-3%; [1,1,2,3,3-2H5]glycerol: 7%.
Glucoregulatory hormonesInsulin, glucagon, cortisol, norepinephrine and epinephrine were measured as described
earlier (22).
proefschrift Soeters.indb 61 4-8-2008 9:13:19
62
Plasma and muscle acylcarnitine measurementsAC plasma concentrations were analyzed as described previously (23). Muscle biopsies
were freeze dried and analyzed as described previously (24). Total long-chain ACs represent
the sum of C12:1-, C12-, C14:2-, C14:1-, C14-, C16:1-, C16-, C18:2-, C18:1- and C18-carnitine.
Calculations and statisticsEndogenous glucose production (EGP) and rate of glucose disposal (Rd) were calculated
using the modified forms of the Steele Equations as described previously (4;19;25). EGP
and Rd were expressed as μmol/kg·min. Lipolysis (glycerol turnover) was calculated by
using formulas for steady state kinetics adapted for stable isotopes (32). Lipolysis was
expressed as μmol/kg·min and as μmol/kcal as proposed by Koutsari et al (30).
Resting energy expenditure (REE) was expressed as kcal/day. FAO and CHO rates were
calculated from O2 consumption and CO2 production (26).
Statistical comparisons and correlation analyses were performed with the Wilcoxon
Signed Rank test and Spearman’s rank correlation coefficient (ρ) respectively. The SPSS
statistical software program version 12.0.2 (SPSS Inc, Chicago, IL) was used for statistical
analysis. Data are presented as median [minimum - maximum].
Results
Anthropometric characteristicsThe subject characteristics have been reported earlier (4): In sum, subject characteristics
were: age: 23 [20 - 26] yrs; weight 70.1 [62.5 – 75.5] kg after 14 h and 69.0 [60.0 –
72.8] kg after 62 h of fasting, P = 0.012; BMI 20.9 [19.2 – 23.3] kg/m2 after 14 h and
20.3 [18.3 – 22.6] kg/m2 after 62 h of fasting, P = 0.011.
Indirect calorimetryREE (kcal/day) in the basal state was significantly higher after 62 h of fasting compared to
14 h of fasting; 1682 [1518 - 1820] kcal/day vs. 1578 [1308 - 1783] kcal/day respectively,
P = 0.017 (Figure 1). During the clamp however, REE was significantly lower after 62 h of
fasting compared to 14 h of fasting; 1604 [1470 - 1734] kcal/day vs. 1815 [1494 - 1933]
kcal/day respectively, P = 0.012.
FAO (μmol/kg·min) in the basal state was significantly higher after 62 h of fasting
compared to 14 h of fasting (Figure 1). During the clamp, FAO remained significantly
higher after 62 h of fasting compared to 14 h of fasting. The absolute change during the
proefschrift Soeters.indb 62 4-8-2008 9:13:19
Muscle fatty acid oxidation in fasting
63
Chapter 4
clamps in FAO was not different between 62
and 14 h of fasting; -5.1 [-6.6 - -2.7] μmol/
kg•min vs. -3.9 [-4.8 - -1.0] μmol/kg•min
respectively, P = 0.8. CHO was significantly
lower after 62 h of fasting compared to 14 h
of fasting (Figure 1). During the clamp, CHO
was significantly lower after 62 h of fasting
compared to 14 h of fasting. The absolute
change between the basal state and the
clamp in CHO was not different between 62
and 14 h of fasting: 12.8 [5.9 - 17.1] μmol/
kg•min vs. 13.6 [7.5 - 19.3] μmol/kg•min
respectively, P = 0.2. If FAO and CHO were
expressed as μmol/kg LBM•min comparable
results were obtained (data not shown).
Glucose and lipid metabolism measurementsPlasma glucose concentrations and EGP
(Table 1) were significantly lower after 62h
of fasting compared to 14 h fasting (4). No
differences were found in plasma glucose
Figure 1, panel A Resting energy expenditure (REE) in the basal state and during the clamp after 14 h (open box plots) and 62 h (grey boxplots) of fasting. *P = 0.017 and 0.012 respectively. **P = 0.017 for the increase and 0.05 for the decrease of REE during the clamps after 62 and 14 h of fasting respectively.Figure 1, panel B Fat oxidation in the basal state and during the clamp after 14 h (open box plots) and 62 h (grey boxplots) of fasting. *P = 0.012 and 0.017 respectively. **P = 0.012 for the decrease of fat oxidation during both clamps after 62 and 14 h of fasting.Figure 1, panel C CHO in the basal state and during the clamp after 14 h (open box plots) and 62 h (grey boxplots) of fasting, *P = 0.012. **P = 0.012 for the increase of CHO during both clamps after 62 and 14 h of fasting.
proefschrift Soeters.indb 63 4-8-2008 9:13:19
64
concentrations after 62h vs 14 h of fasting during the clamp. Rd during the clamp was
significantly lower after 62 h of fasting compared to 14 h of fasting (Table 1).
Basal plasma FFA and rate of lipolysis were significantly higher after 62 h of fasting
(Table 1). Plasma FFA were suppressed during the hyperinsulinaemic euglycaemic clamps
after 62 h and 14 h of fasting (Table 1). Lipolysis was not different between 62 h of
fasting andcompared to 14 h of fasting during the clamp (Table 1).
Glucoregulatory hormonesInsulin was significantly lower after 62 h of fasting in the basal state as well as during
the clamp (Table 1). Other data on glucoregulatory hormones have been presented
elsewhere (4).
Muscle and plasma acylcarnitines in the basal stateAfter 62 h of fasting, muscle free carnitine (FC) was higher compared to 14 h; 9503
[5326 -14360] pmol/mg dry weight vs. 5299 [3331 - 7051] pmol/mg dry weight, P =
0.028. In the basal state muscle long-chain ACs were not significantly different after 62
h vs. 14 h of fasting (Figure 2).
During the clamp muscle FC was not different after 62 h of fasting compared to 14
h of fasting; 1175 [411 - 3380] pmol/mg dry weight vs. 697 [269 - 1149] pmol/mg dry
weight, P = 0.75. Total muscle long-chain ACs were significantly higher during the clamp
after 62 h of fasting compared to 14 h of fasting (Figure 2). Individual muscle long-chain
ACs showed a similar pattern (see Table 2). Total plasma long-chain ACs were significantly
Table 1 Glucose and lipid metabolism measurements
basal state hyperinsulinaemic euglycaemic clamp
14h(n = 8)
62h(n = 8)
P 14h(n = 8)
62h(n = 8)
P
Glucose (mmol/liter) 5.0 [4.5 - 5.6] 3.7 [3.3 - 4.1] 0.011 5.1 [4.9 - 5.3] 4.9 [4.5 - 5.1] 0.107
EGP (μmol/kg•min) 11.8 [8.9 - 19.7] 8.3 [7.0 - 8.7] 0.012 -* -* -
Rd (μmol/kg•min) - - - 60.5 [45.1 - 79.3] 44.4 [37.5 - 51.7] 0.018
FFA (mmol/liter) 0.33 [0.19 - 0.95] 1.10 [0.85 - 1.24] 0.015 <0.02 <0.02 -
Lipolysis (μmol/kg•min) 1.5 [1.1 - 4.4] 3.8 [2.3 - 4.2] 0.017 0.6 [0.2 - 0.8] 0.7 [0.2 - 1.8] 0.249
Lipolysis (μmol/kcal) 101 [68 - 261] 214 [135 - 250] 0.036 32 [11 - 48] 44 [13 - 108] 0.249
Insulin (pmol/liter) 30 [18 - 58] 15 [15 - 20] 0.012 642 [416 - 715] 533 [383 - 663] 0.012
Data are presented as median [minimum - maximum]. EGP, endogenous glucose production; Rd, rate of disposal; FFA, free fatty acids. * During the clamps, EGP and FFA were completely suppressed.
proefschrift Soeters.indb 64 4-8-2008 9:13:20
Muscle fatty acid oxidation in fasting
65
Chapter 4
higher after 62 h of fasting compared to 14 h of fasting in the basal state and during the
clamp. The same was true for individual plasma long-chain ACs (data not shown).
Muscle long-chain acylcarnitines did not correlate with lipolysis, Rd, FAO or plasma
long-chain acylcarnitines (data not shown).
Table 1 Glucose and lipid metabolism measurements
basal state hyperinsulinaemic euglycaemic clamp
14h(n = 8)
62h(n = 8)
P 14h(n = 8)
62h(n = 8)
P
Glucose (mmol/liter) 5.0 [4.5 - 5.6] 3.7 [3.3 - 4.1] 0.011 5.1 [4.9 - 5.3] 4.9 [4.5 - 5.1] 0.107
EGP (μmol/kg•min) 11.8 [8.9 - 19.7] 8.3 [7.0 - 8.7] 0.012 -* -* -
Rd (μmol/kg•min) - - - 60.5 [45.1 - 79.3] 44.4 [37.5 - 51.7] 0.018
FFA (mmol/liter) 0.33 [0.19 - 0.95] 1.10 [0.85 - 1.24] 0.015 <0.02 <0.02 -
Lipolysis (μmol/kg•min) 1.5 [1.1 - 4.4] 3.8 [2.3 - 4.2] 0.017 0.6 [0.2 - 0.8] 0.7 [0.2 - 1.8] 0.249
Lipolysis (μmol/kcal) 101 [68 - 261] 214 [135 - 250] 0.036 32 [11 - 48] 44 [13 - 108] 0.249
Insulin (pmol/liter) 30 [18 - 58] 15 [15 - 20] 0.012 642 [416 - 715] 533 [383 - 663] 0.012
Data are presented as median [minimum - maximum]. EGP, endogenous glucose production; Rd, rate of disposal; FFA, free fatty acids. * During the clamps, EGP and FFA were completely suppressed.
Figure 2, panel A Total long chain plasma acylcarnitines (ACs) the basal state and during the clamp after 14 h (open box plots) and 62 h (grey boxplots) of fasting, *P = 0.012. **P = 0.012 for the decrease of ACs during both clamps after 62 and 14 h of fasting.Figure 2, panel B Total long chain muscle acylcarnitines (ACs) the basal state and during the clamp after 14 h (open box plots) and 62 h (grey boxplots) of fasting, *P = 0.012 for the difference in ACs during the clamp between 62 and 14 h of fasting. ***P = 0.05 for the decrease in ACs between basal state and clamp after 14 h of fasting.
proefschrift Soeters.indb 65 4-8-2008 9:13:20
66
Discussion
In the present study, muscle long-chain acylcarnitines after 14 vs. 16 h of fasting were
studied and correlated to glucose and fatty acid oxidation rates as well as peripheral
insulin sensitivity. In the basal state, there was no significant difference in muscle long-
chain ACs between 14 and 62 h of fasting. In contrast during hyperinsulinemia we found
a decrease in muscle long-chain ACs after 14 h of fasting versus no changes in muscle
long-chain ACs after 62 h of fasting was found. The latter finding was accompanied by
higher whole body FAO.
REE increased approximately 7.5 % during 62 h of fasting, which is in line with
previous observations (3;27-29), although the increase in REE has not fully been
accounted for. It was proposed that increased energy requirements of gluconeogenesis
and ketogenesis are reflected in increased REE (27;28). On the other hand it was
suggested that the norepinephrine induced thermogenic response results in a slight
increase in REE during short-term fasting (27;29), but we did not detect differences in
plasma norepinephrine (4).
Our study confirmed earlier reports on increased plasma FFA and lipolysis (1). Higher
plasma FFA are also found in different models of insulin resistance and are thought to be
one of the main mediators of obesity-induced insulin resistance (30). Lipid mediators that
induce insulin resistance are mainly derived from long-chain fatty acids (31). Moreover,
Table 2. Acylcarnitine concentrations in muscle (pmol/mg dry weight) in the basal state after 14 and 62 h of fasting
basal state hyperinsulinemic euglycemic clamp
Acylcarnitine 14 h 62 h Pb 14 h 62 h Pc Pd Pe
C12:1 1.21 [0.13 - 4.27] 1.17 [0.14 - 2.97] 0.78 0.15 [0 - 1.25] 0.71 [0.21 - 2.63] 0.012 0.025 0.26
C12 4.15 [0.13 - 14.69] 2.42 [0.42 - 7.97] 0.48 0.26 [0.18 - 6.41] 1.45 [0.21 - 5.38] 0.16 0.069 0.48
C14:2 2.05 [0.13 - 9.24] 1.19 [0.31 - 5.51] 0.58 0.15 [0 - 3.59] 0.97 [0.11 - 4.00] 0.036 0.050 0.40
C14:1 7.6 [0.25 - 37.68] 4.68 [0.56 - 20.00] 0.89 0.17 [0.12 - 10.94] 3.41 [0.32 - 9.00] 0.036 0.036 0.33
C14 6.12 [0.25 - 35.31] 4.15 [0.51 - 16.61] 0.67 0.22 [0.12 - 13.44] 2.77 [0.32 - 9,75] 0.16 0.12 0.40
C16:1 6.18 [0.13 - 46.45] 4.59 [0.61 - 22.80] 1.00 0.30 [0.13 - 14.92] 2.48 [0.21 - 14.00] 0.069 0.093 0.67
C16 10.07 [0.88 - 72.80] 6.46 [2.50 - 29.41] 0.67 1.38 [0.42 - 20.15] 9.41 [1.60 - 33.13] 0.012 0.025 0.67
C18:2 3.42 [0.25 - 19.91] 1.84 [0.56 - 6.44] 0.58 0.71 [0.23 - 5.39] 2.40 [0.32 - 12.63] 0.012 0.036 0.78
C18:1 8.10 [0.25 - 88.86] 7.42 [2.36 - 37.12] 0.67 1.56 [0.58 - 20.00] 6.18 [0.96 - 32.63] 0.012 0.036 0.48
C18 3.46 [0.25 - 25.36] 2.32 [0.97 - 7.29] 0.40 1.18 [0.47 - 5.39] 3.04 [0.85 - 9.50] 0.012 0.017 0.33
Data are presented as median [minimum - maximum]. a N = 6. b,cP-values represent differences between basal states and clamps after 14 and 62 h of fasting respectively.d,eP-values represent differences within clamps after 14 and 62 h respectively.
proefschrift Soeters.indb 66 4-8-2008 9:13:20
Muscle fatty acid oxidation in fasting
67
Chapter 4
muscle long-chain ACs have been suggested to induce peripheral insulin resistance in
animal studies (8;11;12). Since fasting induces insulin-resistance (2;4), we hypothesized
that this may be related to an increase in muscle long-chain ACs. Human data on muscle
ACs during fasting are lacking but previous animal studies showed an increase of muscle
long-chain ACs in rodents after fasting (8;17).
The lack of an increase in muscle long-chain ACs between 14 and 62 h of fasting is
unexpected since we demonstrated an increase in whole body FAO. The CPT1 dependent
rate of long-chain fatty acid entrance into the mitochondria is thought to determine the
rate of FAO (5;7). Our data imply that the muscle concentration of ACs during fasting in
humans does not reflect the fatty acid oxidation flux.
Hyperinsulinemia resulted in a decreased concentration of long-chain muscle ACs after
14 h of fasting. This was in line with the lower whole body FAO during the clamp after
14 h of fasting. After 62 h of fasting, the suppressive effect of insulin on muscle long-
chain ACs was not found. Increased muscle long- chain ACs during the clamp after 62 h
of fasting are unlikely to reflect accumulation of non utilizable long-chain ACs or ongoing
FAO since we could not demonstrate such a relationship in the basal state after 62 h of
fasting. However, ongoing FAO is likely to occur since clamp values of whole body FAO
were higher after 62 h of fasting compared to 14 h of fasting. This suggests that despite
5 h of hyperinsulinemia, peripheral glucose uptake is still attenuated and activated fatty
acids are continued to be transported to the mitochondrion in order to be oxidized (5;7).
Table 2. Acylcarnitine concentrations in muscle (pmol/mg dry weight) in the basal state after 14 and 62 h of fasting
basal state hyperinsulinemic euglycemic clamp
Acylcarnitine 14 h 62 h Pb 14 h 62 h Pc Pd Pe
C12:1 1.21 [0.13 - 4.27] 1.17 [0.14 - 2.97] 0.78 0.15 [0 - 1.25] 0.71 [0.21 - 2.63] 0.012 0.025 0.26
C12 4.15 [0.13 - 14.69] 2.42 [0.42 - 7.97] 0.48 0.26 [0.18 - 6.41] 1.45 [0.21 - 5.38] 0.16 0.069 0.48
C14:2 2.05 [0.13 - 9.24] 1.19 [0.31 - 5.51] 0.58 0.15 [0 - 3.59] 0.97 [0.11 - 4.00] 0.036 0.050 0.40
C14:1 7.6 [0.25 - 37.68] 4.68 [0.56 - 20.00] 0.89 0.17 [0.12 - 10.94] 3.41 [0.32 - 9.00] 0.036 0.036 0.33
C14 6.12 [0.25 - 35.31] 4.15 [0.51 - 16.61] 0.67 0.22 [0.12 - 13.44] 2.77 [0.32 - 9,75] 0.16 0.12 0.40
C16:1 6.18 [0.13 - 46.45] 4.59 [0.61 - 22.80] 1.00 0.30 [0.13 - 14.92] 2.48 [0.21 - 14.00] 0.069 0.093 0.67
C16 10.07 [0.88 - 72.80] 6.46 [2.50 - 29.41] 0.67 1.38 [0.42 - 20.15] 9.41 [1.60 - 33.13] 0.012 0.025 0.67
C18:2 3.42 [0.25 - 19.91] 1.84 [0.56 - 6.44] 0.58 0.71 [0.23 - 5.39] 2.40 [0.32 - 12.63] 0.012 0.036 0.78
C18:1 8.10 [0.25 - 88.86] 7.42 [2.36 - 37.12] 0.67 1.56 [0.58 - 20.00] 6.18 [0.96 - 32.63] 0.012 0.036 0.48
C18 3.46 [0.25 - 25.36] 2.32 [0.97 - 7.29] 0.40 1.18 [0.47 - 5.39] 3.04 [0.85 - 9.50] 0.012 0.017 0.33
Data are presented as median [minimum - maximum]. a N = 6. b,cP-values represent differences between basal states and clamps after 14 and 62 h of fasting respectively.d,eP-values represent differences within clamps after 14 and 62 h respectively.
proefschrift Soeters.indb 67 4-8-2008 9:13:20
68
Ongoing FAO may be needed since peripheral glucose uptake is attenuated. Our data on
increased plasma long-chain ACs after 62 h of fasting confirm older studies (13-15). In
contrast to our finding in muscle, plasma long-chain ACs were higher after 62 compared
to 14 h of fasting, but decreased equally during hyperinsulinemia. This, and the absence
of a correlation of plasma ACs with muscle ACs negates that plasma ACs reflect muscle
ACs. It may support the notion that the liver is the most likely source of plasma long-chain
ACs during short-term fasting (7;32;33).
Muscle FC increased during 62 h compared to 14 of fasting which reflects the
dependence of FAO on FC for transport of long-chain acyl-CoAs across the mitochondrial
membranes.
Although we found lower insulin levels during the clamp after 62 h of fasting, it is not
likely that these have interfered with our results as discussed earlier (4).
In conclusion we show that fasting for 62 h results in higher rates of lipolysis and
FAO together with lower CHO rates and lower peripheral insulin sensitivity. These flux
rates are not paralleled with an increase in muscle long-chain ACs after 62 h of fasting in
the basal state. However, during hyperinsulinemia, the suppression of the concentration
of muscle long-chain ACs was less compared to after 14 h of fasting. Also, FAO rates
remained higher during the clamp after 62 h of fasting. Despite earlier reports on possible
interference of the muscle long-chain ACs with peripheral insulin sensitivity, our study
does not support such a role for muscle long-chain ACs during fasting. To clarify whether
muscle ACs are just innocent bystanders or active players in insulin resistance, further
studies in different models of insulin resistance are needed.
AcknowledgementsWe thank A.F.C Ruiter and M. Doolaard for excellent assistance on laboratory analyses of
stable isotopes and acylcarnitines respectively.
Reference List 1. Klein S, Sakurai Y, Romijn JA, Carroll RM. Progressive alterations in lipid and glucose
metabolism during short-term fasting in young adult men. Am J Physiol Endocrinol Metab 1993; 265(5):E801-E806.
2. Bjorkman O, Eriksson LS. Influence of a 60-hour fast on insulin-mediated splanchnic and peripheral glucose metabolism in humans. J Clin Invest 1985; 76(1):87-92.
3. Bergman BC, Cornier MA, Horton TJ, Bessesen DH. Effects of fasting on insulin action and glucose kinetics in lean and obese men and women. Am J Physiol Endocrinol Metab 2007; 293(4):E1103-E1111.
proefschrift Soeters.indb 68 4-8-2008 9:13:20
Muscle fatty acid oxidation in fasting
69
Chapter 4
4. Soeters MR, Sauerwein HP, Dubbelhuis PF et al. Muscle adaptation to short-term fasting in healthy lean humans. J Clin Endocrinol Metab 2008;jc.
5. Stephens FB, Constantin-Teodosiu D, Greenhaff PL. New insights concerning the role of carnitine in the regulation of fuel metabolism in skeletal muscle. J Physiol 2007; 581(2):431-444.
6. Ramsay RR, Gandour RD, van der Leij FR. Molecular enzymology of carnitine transfer and transport. Biochim Biophys Acta 2001; 1546(1):21-43.
7. Eaton S. Control of mitochondrial [beta]-oxidation flux. Progress in Lipid Research 2002; 41(3):197-239.
8. Koves TR, Ussher JR, Noland RC et al. Mitochondrial overload and incomplete fatty acid oxidation contribute to skeletal muscle insulin resistance. Cell Metab 2008; 7(1):45-56.
9. Bartlett K, Pourfarzam M. Tandem mass spectrometry GÇö The potential. Journal of Inherited Metabolic Disease 1999; 22(4):568-571.
10. Bartlett K, Eaton SJ, Pourfarzam M. New developments in neonatal screening. Arch Dis Child Fetal Neonatal Ed 1997; 77(2):F151-F154.
11. Muoio DM, Koves TR. Lipid-induced metabolic dysfunction in skeletal muscle. Novartis Found Symp 2007; 286:24-38; discussion 38-46, 162-3,;%196-203.:24-38.
12. An J, Muoio DM, Shiota M et al. Hepatic expression of malonyl-CoA decarboxylase reverses muscle, liver and whole-animal insulin resistance. Nat Med 2004; 10(3):268-274.
13. Hoppel CL, Genuth SM. Carnitine metabolism in normal-weight and obese human subjects during fasting. Am J Physiol 1980; 238(5):E409-E415.
14. Frohlich J, Seccombe DW, Hahn P, Dodek P, Hynie I. Effect of fasting on free and esterified carnitine levels in human serum and urine: Correlation with serum levels of free fatty acids and [beta]-hydroxybutyrate. Metabolism 1978; 27(5):555-561.
15. Costa CCG, De Almeide IT, Jakobs Corn, Poll-The B-T, Durna M. Dynamic Changes of Plasma Acylcarnitine Levels Induced by Fasting and Sunflower Oil Challenge Test in Children. [Miscellaneous Article]. Pediatric Research 1999; 46(4):440.
16. Erfle JD, Sauer F. Acetyl Coenzyme A and Acetylcarnitine Concentration and Turnover Rates in Muscle and Liver of the Ketotic Rat and Guinea Pig. J Biol Chem 1967; 242(9):1988-1996.
17. Kerner J, Bieber LL. The effect of electrical stimulation, fasting and anesthesia on the carnitine(s) and acyl-carnitines of rat white and red skeletal muscle fibres. Comp Biochem Physiol B 1983; 75(2):311-316.
18. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 2003; 26(90001):5S-20.
19. Finegood DT, Bergman RN, Vranic M. Estimation of endogenous glucose production during hyperinsulinemic-euglycemic glucose clamps. Comparison of unlabeled and labeled exogenous glucose infusates. Diabetes 1987; 36(8):914-924.
20. Ackermans MT, Pereira Arias AM, Bisschop PH, Endert E, Sauerwein HP, Romijn JA. The quantification of gluconeogenesis in healthy men by (2)H2O and [2-(13)C]glycerol yields different results: rates of gluconeogenesis in healthy men measured with (2)H2O are higher than those measured with [2-(13)C]glycerol. J Clin Endocrinol Metab 2001; 86(5):2220-2226.
21. Ackermans MT, Ruiter AF, Endert E. Determination of glycerol concentrations and glycerol isotopic enrichments in human plasma by gas chromatography/mass spectrometry. Anal Biochem 1998; 258(1):80-86.
22. Soeters MR, Sauerwein HP, Groener JE et al. Gender-Related Differences in the Metabolic Response to Fasting. J Clin Endocrinol Metab 2007; 92(9):3646-3652.
23. Vreken P, van Lint AE, Bootsma AH, Overmars H, Wanders RJ, van Gennip AH. Quantitative plasma acylcarnitine analysis using electrospray tandem mass spectrometry for the diagnosis of organic acidaemias and fatty acid oxidation defects. J Inherit Metab Dis 1999; 22(3):302-306.
24. van Vlies N, Tian L, Overmars H et al. Characterization of carnitine and fatty acid metabolism in the long-chain acyl-CoA dehydrogenase-deficient mouse. Biochem J 2005; 387(1):185-193.
proefschrift Soeters.indb 69 4-8-2008 9:13:20
70
25. Steele R. Influences of glucose loading and of injected insulin on hepatic glucose output. Ann N Y Acad Sci 1959; 82:420-30.:420-430.
26. Frayn KN. Calculation of substrate oxidation rates in vivo from gaseous exchange. J Appl Physiol 1983; 55(2):628-634.
27. Webber J, Taylor J, Greathead H, Dawson J, Buttery PJ, Macdonald IA. Effects of fasting on fatty acid kinetics and on the cardiovascular, thermogenic and metabolic responses to the glucose clamp. Clin Sci (Lond) 1994; 87(6):697-706.
28. Mansell PI, Macdonald IA. The effect of starvation on insulin-induced glucose disposal and thermogenesis in humans. Metabolism 1990; 39(5):502-510.
29. Zauner C, Schneeweiss B, Kranz A et al. Resting energy expenditure in short-term starvation is increased as a result of an increase in serum norepinephrine. Am J Clin Nutr 2000; 71(6):1511-1515.
30. Boden G. Interaction between free fatty acids and glucose metabolism. Curr Opin Clin Nutr Metab Care 2002; 5(5):545-549.
31. Holland WL, Knotts TA, Chavez JA, Wang LP, Hoehn KL, Summers SA. Lipid Mediators of Insulin Resistance. Nutrition Reviews 2007; 65:39-46.
32. Bartlett K, Bhuiyan AK, ynsley-Green A, Butler PC, Alberti KG. Human forearm arteriovenous differences of carnitine, short-chain acylcarnitine and long-chain acylcarnitine. Clin Sci (Lond) 1989; 77(4):413-416.
33. Eaton S, Bartlett K, Pourfarzam M, Markley MA, New KJ, Quant PA. Production and export of acylcarnitine esters by neonatal rat hepatocytes. Adv Exp Med Biol 1999; 466:155-159.
proefschrift Soeters.indb 70 4-8-2008 9:13:20
proefschrift Soeters.indb 71 4-8-2008 9:13:20
proefschrift Soeters.indb 72 4-8-2008 9:13:20
Muscle D-3-hydroxybutyrylcarnitine: an alternative pathway in ketone body metabolism5Maarten R. Soeters1, Hans P. Sauerwein1,
Marinus Duran2, Jos P. Ruiter2, Mariëtte T.
Ackermans3, Paul E. Minkler4, Charles L.
Hoppel4,5, Ronald J. Wanders2, Sander M.
Houten2 and Mireille J. Serlie1
1Department of Endocrinology and Metabolism2Department of Clinical Chemistry, Laboratory
Genetic Metabolic Diseases3Department of Clinical Chemistry, Laboratory
of Endocrinology
Academic Medical Center, University of
Amsterdam, Amsterdam, the Netherlands4Case Western Reserve University, School of
Medicine, Department of Medicine, Cleveland,
United States5Case Western Reserve University, School
of Medicine, Department of Pharmacology,
Cleveland, United States
Submitted
proefschrift Soeters.indb 73 4-8-2008 9:13:20
Abstract
Background: D-3-hydroxybutyrate (D-3HB) and acetoacetate play an important role
in the adaptation to fasting. It has been suggested by at least two separate studies
that D-3HB can be coupled to carnitine to form 3-hydroxybutyrylcarnitine. In contrast to
L-hydroxybutyryl-CoA that can be coupled to carnitine resulting in the formation of L-3HB-
carnitine, it is unknown whether and how D-3HB can be coupled to carnitine resulting in
D-3HB-carnitine in humans.
Study-aim: To assess which stereo isomers of 3-hydroxybutyrylcarnitine are present in
vivo.
Methods: 12 lean healthy men underwent a 38 h fasting period to induce ketosis. D-3HB
kinetics and stereo isomers of muscle 3-hydroxybutyrylcarnitine were measured. Studies in
mouse liver and muscle were performed to explore synthesis of D-3HB-carnitine, focusing
on a hypothetical acyl-CoA synthetase (ACS) and succinyl-CoA oxoacid transferase (SCOT)
pathway.
Results: Muscle D-3HB-carnitine was approximately 7.5 fold higher compared to L-3HB-
carnitine. Muscle D-3HB-carnitine and D-3-HB turnover correlated significantly. The ACS
pathway was active in liver and muscle homogenates though less than SCOT pathway
that was only active in muscle homogenates.
Conclusions: We show that D-3HB-carnitine can be formed in muscle by mitochondrial
SCOT and carnitine acyltransferase. Furthermore muscle D-3HB-carnitine correlates with
the plasma turnover of D-3HB in lean healthy men during ketosis. Our data provide a
newly identified alternative pathway of ketone body metabolism. The purpose of D-3HB-
carnitine synthesis and its fate remain to be elucidated.
proefschrift Soeters.indb 74 4-8-2008 9:13:21
D-3-hydroxybutyrylcarnitine: an alternative ketone body pathw
ay
75
Chapter 5
Introduction
The ketone bodies (KBs) D-3-hydroxybutyrate (D-3HB) and acetoacetate (AcAc) play an
important role in the adaptation to fasting, serving as an important fuel for the central
nervous system (1). KBs are produced mainly from fatty acids that are degraded through
beta-oxidation within liver mitochondria. The resulting acetyl-CoA enters the tricarboxylic
acid cycle (TCAC) or is channelled into the ketogenesis pathway (2;3). Via passive diffusion
and active transport (monocarboxylate transporters 1, 2 and 4) D-3HB and AcAc will
reach distant target tissues (e.g. central nervous system and skeletal muscle). D-3HB is
oxidized to AcAc (2-4). AcAc then is coupled to CoA by succinyl-CoA 3-ketoacid(oxoacid)
transferase (SCOT). The resulting acetoacetyl-CoA will yield two acetyl-CoAs that can be
oxidized in the TCAC.
It has been suggested by at least two separate studies that D-3HB can be coupled
to carnitine to form 3-hydroxybutyrylcarnitine (D-3HB-carnitine) (5;6). However, this
raises a few questions. Although it has been described that D-3HB can be activated to
D-3HB-CoA in rat liver and hepatoma cells, this has not been established in humans (7-9).
Furthermore, it is unknown whether and how D-3HB-CoA can be coupled to carnitine
resulting in D-3HB-carnitine.
This is in contrast to its generally known stereo isomer L-hydroxybutyryl-CoA (L-3HB-
CoA), formed via the second step in the beta-oxidation of butyryl-CoA (10) i.e. hydration
of crotonyl-CoA, thereby yielding the stereo-isomer L-3HB-CoA, which can be coupled to
carnitine resulting in the formation of L-3HB-carnitine.
Although it was proposed in the earlier mentioned studies that the total amount
of tissue hydroxybutyrylcarnitine was derived from D-3HB and carnitine, the quantative
contributions of the L and D stereo isomers were not accounted for in these studies (5;6).
This leaves the question which stereo isomers are present in vivo, unanswered.
In this study we investigated the relation between total D-3-HB turnover and muscle
hydroxybutyryl-carnitine in 12 lean healthy men who underwent a 38 h fasting period to
induce ketosis. D-3HB kinetics were measured using stable isotopes and the pancreatic
clamp technique. Furthermore we assessed the quantitative contribution of the D- and
L-stereo isomers to the total amount of muscle hydroxybutyrylcarnitine in the basal
state, i.e. after 38 h of fasting and during a pancreatic clamp. The latter to study the
changes in muscle D and L-3HB-carnitine upon a decrease in ketogenesis induced
by mild hyperinsulinemia. We hypothesized total KB turnover rates to correlate with
muscle D-3HB-carnitine. Additionally, we performed studies in mouse liver and muscle
proefschrift Soeters.indb 75 4-8-2008 9:13:21
76
homogenates to explore potential metabolic pathways responsible for the synthesis of
D-3HB-carnitine; here we focused on SCOT and acyl-CoA synthetase (ACS). The activation
of D-3HB and subsequent conversion to acylcarnitine would be a novel concept, offering
an alternative pathway in ketone body metabolism.
Experimental procedures
Human pancreatic clamp studiesData were included of twelve healthy lean male subjects who participated as controls in
a study on obesity induced insulin resistance and KB metabolism (age: 24 (20 - 53) yrs;
weight 73.8 (59.0 – 83.5); BMI 22.4 (18.9 – 24.7) kg/m2) (Soeters, unpublished data).
Criteria for inclusion were 1) absence of a family history of diabetes; 2) age 18–60 yr; 3)
BMI 18-25 kg/m2; 4) normal oral glucose tolerance test (OGTT) according to American
Diabetes Association-criteria (11); 5) no excessive sport activities, i.e. < 3 times per week;
and 6) no medication. Subjects were in self reported good health, confirmed by medical
history, routine laboratory and physical examination. Written informed consent was
obtained from all subjects after explanation of purposes, nature, and potential risks of
the study. The study was approved by the Medical Ethical Committee of the Academic
Medical Center of the University of Amsterdam.
Subjects started fasting at 2000 h two days before the study day to induce ketosis. On
the study day, volunteers arrived at the metabolic unit of the Academic Medical Center at
0730 h. D[2,4-13C2]-3HB (>99% enriched; Cambridge Isotopes, Andover, USA) was used
to study KB turnover. At T = 0 h (0800 h), blood samples were drawn for determination
of background enrichments and a primed continuous infusion of D[2,4-13C2]-3HB (prime,
1.0 μmol/kg; continuous, 0.10 μmol/kg·min) was started and continued until the end
of the study. After two hours of equilibration (38 h of fasting), 3 blood samples were
drawn for KB enrichments and concentrations and 1 for glucoregulatory hormones
and FFA. Then a pancreatic clamp was started at T = 2.5. Although the clamp included
two steps (low dose insulin infusion at a rate of 4.5mU/m2·min and 7.5mU/m2·min
respectively (Actrapid 100 IU/ml; Novo Nordisk Farma B.V., Alphen aan den Rijn, the
Netherlands), only the last step is discussed in this report (referred to as “the clamp”).
Glucose 5% was infused to maintain plasma glucose levels of 5 mmol/L. At the same time
(T=2.5 h) infusion of somatostatin (250 μg/h; Somatostatine-ucb; UCB Pharma, Breda,
the Netherlands) was started, as well as a glucagon infusion (1 ng/kg·min); Glucagen;
Novo Nordisk, Alphen aan den Rijn, the Netherlands) to replace endogenous glucagon
proefschrift Soeters.indb 76 4-8-2008 9:13:21
D-3-hydroxybutyrylcarnitine: an alternative ketone body pathw
ay
77
Chapter 5
concentrations throughout the clamp. Plasma glucose levels were monitored in 10 minute
intervals. At the end of the clamp 5 blood samples were drawn for KB enrichments and
concentrations, plasma glucose levels and 1 for glucoregulatory hormones and FFA.
Muscle biopsies were performed at the end of both basal state and clamp as described
earlier (12).
Human plasma D-3HB kinetics and FFAD-3HB samples were drawn in chilled sodium-fluoride tubes and directly deproteinized with
ice cold perchloric acid 6% (1:1). D-3HB was measured spectrophotometrically using D-3-
hydroxy-butyrate dehydrogenase (COBAS-FARA centrifugal analyzer, Roche Diagnostics,
Almere, the Netherlands). D-3HB tracer tracee ratio’s (TTR) were measured as follows:
After centrifugation of the acidified KB samples, the supernatants were neutralized
with KOH and HClO4. Following centrifugation the supernatants were acidified with
HCl and extracted twice with ethylacetate. The combined extracts were dried under
nitrogen at room temperature. D-3HB was converted to its t-butyldimethylsilyl derivative
with MTBSTFA and pyridine. The sample was injected into an Agilent 6890/5973
MSD gaschromotograph/mass spectrometer system (Agilent Technologies, Palo Alto,
CA). Separation was achieved on a Varian (Middelburg, The Netherlands) CP-SIL 19CB
column (30m x 0.25mm x 0.15μm). Selected ion monitoring (EI), data acquisition and
quantitative calculations were performed using the Agilent Chemstation software. Ions
were monitored at m/z 275 for unlabeled D-3HB and m/z 277 for D[2,4-13C2]-3HB.
The enrichment in D-3HB was determined from standard curves of known enrichments
injected in the same run.
D-3HB turnover (Ra) was calculated as the rate of isotope infusion (μmol/kg·min)
divided by plateau tracer tracee ratios (TTR), hence D-3HB Ra was expressed as μmol/
kg·min. Plasma glucose and FFA were measured as described previously (12). Plasma
insulin levels were measured with an ultra sensitive human insulin RIA kit (Linco HI-11K,
St. Charles, Missouri, USA; Detection limit 1.5 pmol/L).
Human muscle 3HB-carnitine and D- and L-3HB isomer measurementsMuscle hydroxybutyryl-carnitine levels were determined in freeze dried muscle specimens
by tandem MS as described previously (13).
Muscle levels of the stereo isomers D- and L-3HB-carnitine were determined by high
performance liquid chromatography (HPLC) tandem MS with minor modifications as
described by Minkler et al (14).
proefschrift Soeters.indb 77 4-8-2008 9:13:21
78
Statistical analysesAll human data were analyzed with non parametric tests (Wilcoxon Signed Rank test).
Correlations were expressed as Spearman’s rank correlation coefficient (ρ). The SPSS
statistical software program version 14.0 (SPSS Inc, Chicago, IL) was used for statistical
analysis. Data are presented as median [minimum -maximum].
Experimental procedures muscle and liver homogenate studiesMouse muscle (strain C57BL/6) was homogenized in PBS using an ultra turrax followed
by sonication (3 times 40J). Liver was homogenized in PBS using sonication (2 times 40J)
with constant cooling in an ice-water bath (4°C). Homogenates were centrifuged for 1
minute at 20xg. The incubation mixtures for ACS activity included, 100mM TrisHCl pH8,
1mM coenzyme A, 10mM MgCl2, 10mM ATP, 5mM carnitine. The incubation mixtures for
SCOT activity included: 100mM TrisHCl pH8.0, 0.25 mM succinyl-CoA, 5mM carnitine. For
both mixtures (100 μL), we used 10mM D,L-3-hydroxybutyrate, 5mM D-3-hydroxybutyrate
or 5mM L-3-hydroxybutyrate as substrate. The incubations were terminated by adding
acetonitril containing the internal standards. After drying the samples with nitrogen,
the reaction products were butylated using 1-butanol/acetylchloride (4:1). Samples
were dissolved in acetonitril and analyzed by MS/MS. Carnitine acyltransferase activity
was measured using the same tissue homogenates or purified pigeon breast carnitine
acetyltransferase (CAT). The incubation mixtures (100 μL) included 100mM Tris.Cl pH8
and 5mM carnitine. As substrates, we used 220 μM D,L-3-hydroxybutyryl-CoA and 110 μM
D-3-hydroxybutyryl-CoA. The D-3-hydroxybutyryl-CoA was prepared enzymatically from
D,L-3-hydroxybutyryl-CoA, by using purified SCHAD that converts L-3-hydroxybutyryl-CoA
into acetoacetyl-CoA. The D-3-hydroxybutyryl-CoA was purified using HPLC.
RESULTS
Human plasma D-3HB kinetics, FFA, glucose and insulinBasal state values of D-3HB Ra, FFA, glucose and insulin are depicted in Table 1. Low
dose insulin infusion significantly decreased plasma D-3HB levels as well as D-3HB Ra
during the clamp compared to the basal state (Table 1). Likewise, plasma FFA decreased
during the clamp compared to the basal state. Insulin infusion increased insulin levels as
expected during the clamp compared to the basal state.
proefschrift Soeters.indb 78 4-8-2008 9:13:21
D-3-hydroxybutyrylcarnitine: an alternative ketone body pathw
ay
79
Chapter 5
Human muscle 3HB-carnitine correlates with D-3HB RaTotal (D and L isomers) muscle 3-HB-carnitine are depicted in Figure 1, panel A. Total
muscle 3-HB-carnitine levers were lower after the clamp. We found a significant correlation
in the basal state between total muscle 3-HB-carnitine and D-3HB Ra suggesting that the
total amount of muscle 3-HB-carnitine is directly related to the 3-HB flux (Figure 1, panel B).
Plasma FFA were found to correlate with total muscle 3-HB-carnitine (ρ 0.83, P = 0.001).
D-3HB-carnitineTotal muscle 3-HB-carnitine may consist of a D and L stereo isomer and since regular
tandem MS does not discriminate between these two stereo isomers, we used HPLC-
tandem MS to separate the two isomers. We found muscle D-3HB-carnitine to be
approximately 7.5 fold higher compared to L-3HB-carnitine (Figure 1, panel C). Of interest,
L -3-HB-carnitine could not be detected during the pancreatic clamp whereas D -3-HB-
carnitine was significantly lower after the clamp. Results on muscle D -3HB-carnitine were
missing in 2 subjects in the basal state and 3 during the clamp because measurements
were below the detection limit. A significant positive correlation was found between
muscle D-3HB-carnitine and the D-3-HB turnover in the basal state (Figure 1, panel D),
however this was not the case during the clamp (clamp; ρ= 0.46, P = 0.13).
Synthesis of D-3-HB-carnitine in liver and muscle homogenates of miceWe hypothesized that the D-3-HB-carnitine could be formed by two alternative pathways.
The first potential route is via acyl-CoA synthetase (ACS). D-3-HB can be activated to
D-3HB-CoA by ACS activity, followed by the conversion to D-3HB-carnitine by a carnitine
acyltransferase. This enzymatic reaction proceeds after the addition of ATP, free CoA
and carnitine. The second hypothetical pathway is mediated by SCOT. SCOT transfers
the CoA from succinyl-CoA to D-3HB, followed by the conversion to D-3HB-carnitine.
Succinyl-CoA and carnitine are needed for this reaction. Both pathways were assessed
Table 1 Plasma measurements healhty lean volunteers
basal state clamp
D-3HB (mmol/liter) 1.24 [0.30 - 2.03] 0.05 [0 - 0.40] 0.002
D-3HB Ra (μmol/kcal) 11.6 [6.9 - 22.4] 4.7 [1.3 - 9.0] 0.002
FFA (mmol/liter) 1.04 [0.72 - 1.21] 0.22 [0.06 - 0.48] 0.012
Glucose (mmol/liter) 4.1 [3.5 - 4.5] 5.0 [4.5 - 5.2] 0.002
Insulin (pmol/liter) 13 [6 - 27] 49 [26 - 69] 0.002
N = 12. Ra, rate of appearance; FFA, free fatty acids. Data are presented as median [minimum - maximum].
proefschrift Soeters.indb 79 4-8-2008 9:13:21
80
in mouse muscle and liver homogenates by measuring the formation of 3-HB-carnitine
from D- and L-3-HB. This method involves the measurement of two enzymatic steps, i.e.
the activation of the 3-HB to 3-HB-CoA and the subsequent conversion to 3-HB-carnitine.
The ACS pathway was active in liver and muscle homogenates (Figure 2, panels A and B).
L-3-HB was converted into 3-HB-carnitine more efficiently than D-3-HB.
FIGURE 1 Panel A; Total amount of muscle 3-hydroxybutyrylcarnitine in the basal state and during the clamp * P = 0.002 (decrease in D-3-HB-carnitine during the clamp). Boxes and whiskers represent the median, minimum and maximum.Panel B; Correlation of the total amount of muscle 3-hydroxybutyrylcarnitine with the D-3HB Ra in the basal state: ρ = 0.74, P = 0.006.Panel C; Absolute contribution of D-3-HB-carnitine (empty bars) and L-3-HB-carnitine (grey bar) to the total amount of muscle 3-hydroxybutyrylcarnitine in the basal state (n = 10) and during the clamp (n = 9). Bars represent the median.Panel D; Correlation of muscle D-3-HB-carnitine with the D-3HB Ra in the basal state: ρ = 0.69, P = 0.029 (n = 10).
proefschrift Soeters.indb 80 4-8-2008 9:13:21
D-3-hydroxybutyrylcarnitine: an alternative ketone body pathw
ay
81
Chapter 5
proefschrift Soeters.indb 81 4-8-2008 9:13:21
82
Remarkably, the SCOT route was active in muscle homogenates only (Figure 2, panel
C). With the addition of succinyl-CoA, D-3-HB was converted into 3-HB-carnitine more
efficiently than L-3-HB. In liver this route was practically undetectable (Figure 2, panel D),
which is consistent with the absence of SCOT activity in this tissue (3;15).
The second step in the formation of 3-HB-carnitine was hypothesized to be catalyzed
by a carnitine acyltransferase. In liver and muscle homogenates this reaction proceeds
as evident from the above described results. This was further evaluated in experiments
were D-3HB-CoA was added to muscle and liver homogenates yielding D-3HB-carnitine. In
liver homogenates this reaction was more efficient (Figure 2, panels E and F). Finally, we
tested if this reaction is performed by CAT. Purified pigeon breast CAT was able to convert
D-3HB-CoA to D-3-HB carnitine, although with low efficiency (Figure 2, panel G).
DISCUSSION
In this study we examined the relationship between the ketone body turnover and
the presence of muscle D-3HB-carnitine and explored which isomer of 3-HB-carnitine is
predominant in human skeletal muscle during ketosis induced by 38 h of fasting. We
found D-3HB-carnitine to be the predominant 3HB-carnitine. Moreover, we show that
muscle contains the full enzymatic machinery to synthesize D-3-HB-carnitine as shown
in muscle homogenates in mice. This newly identified pathway involves an alternative
metabolic route of ketone bodies in muscle.
FIGURE 2 Panel A; Synthesis of D/L-3HB-carnitine in muscle homogenates of mice in presence of D-3HB, CoA, ATP and carnitine, catalyzed by acyl-CoA synthetase (ACS). D/L-3HB-carnitine are presented as open and closed boxes respectively. Panel B; Synthesis of D/L-3HB-carnitine in liver homogenates of mice in presence of D-3HB, CoA, ATP and carnitine, catalyzed by acyl-CoA synthetase (ACS). D/L-3HB-carnitine are presented as open and closed boxes respectively.Panel C; Synthesis of D/L-3HB-carnitine in muscle homogenates of mice in presence of D-3HB, succinyl-CoA and carnitine, catalyzed by succinyl-CoA 3-ketoacid(oxoacid) transferase (SCOT). D/L-3HB-carnitine are presented as open and closed boxes respectively.Panel D; Synthesis of D/L-3HB-carnitine in liver homogenates of mice in presence of D-3HB, succinyl-CoA and carnitine, catalyzed by succinyl-CoA 3-ketoacid(oxoacid) transferase (SCOT). D/L-3HB-carnitine are presented as open and closed boxes respectively.Panel E; Synthesis of D-3HB-carnitine in muscle homogenates of mice in presence of D-3HB-CoA and carnitine, catalyzed by acyltransferase.Panel F; Synthesis of D-3HB-carnitine in liver homogenates of mice in presence of D-3HB-CoA and carnitine, catalyzed by acyltransferase.Panel G; Synthesis of D-3HB-carnitine in Tris.Cl in presence of D-3HB-CoA and carnitine with purified pigeon carnitine acyltransferase (CAT).
proefschrift Soeters.indb 82 4-8-2008 9:13:21
D-3-hydroxybutyrylcarnitine: an alternative ketone body pathw
ay
83
Chapter 5
We confirmed earlier reports on the association between muscle total 3-HB-carnitine
and plasma KBs and FFA. It was shown by Hack et al that 3-HB-carnitine in adipose tissue
microdialysis fluid in children correlated with plasma KBs after 24 h of fasting (6). Hack et
al suggested that D-3HB could be coupled to carnitine in an unspecified enzymatic reaction
as was observed for acetyl-CoA. An et al performed extensive acylcarnitine profiling in
muscle samples of rodents and found that total muscle 3-HB-carnitine correlated with
plasma FFA levels (5). They proposed muscle ketogenesis to be responsible for the amount
of total muscle 3-HB-carnitine. However both studies did not validate their hypothesis by
measuring both isomers of 3-HB-carnitine.
We used HPLC-MS/MS to separate D- and L-3-HB-carnitine and show for the first time
that most of total muscle 3-HB-carnitine (approximately 88%) consists of D-3HB-carnitine
after 38 h of fasting. Additionally, after 4 h of low dose insulin infusion, total muscle
3-HB-carnitine consisted entirely of D-3-HB-carnitine while the turnover rate of D-3HB
had decreased by approximately 59%. Although we may not have detected low levels of
muscle L-3-HB-carnitine, this suggests that beta-oxidation is decreased, thereby yielding
less muscle L-3HB-carnitine. The metabolic pathway of D-3HB-carnitine is less clear.
Plasma D-3HB during short-term fasting is mainly derived from hepatic KB production
(2;3). Therefore we measured the D-3HB production rate and found a correlation with
muscle D-3HB-carnitine. This suggests that plasma D-3HB is transported to the myocyte
where activation and subsequent coupling to carnitine occurs and fits with the assumption
that the increased muscle D-3HB delivery during fasting exceeds muscle D-3HB oxidation
analogous to what has been described previously for fatty acids in liver and muscle
during fasting (16-18).
The existence of muscle D-3HB-carnitine and its relation to KB production in healthy
lean volunteers prompts the question what metabolic pathway is responsible for the
synthesis of D-3HB-carnitine. At least two reactions are critical: the activation of D-3HB
to its CoA ester and secondly the exchange of CoA for carnitine. With this in mind we
performed additional experiments in mice muscle and liver homogenates. We explored
two possible metabolic pathways that may activate D-3HB: first we investigated whether
D-3HB could be activated by ACS in the presence of ATP and CoA only. In muscle this
was the case to some extend, however L-3HB appeared to be a better substrate yielding
more L-3HB-carnitine (over 3-fold) compared to the D-isomer. Although L-3HB has been
debated to be a natural occurring KB with low concentration (19-21), it was described in
in vitro studies to be activated by a (cytosolic) ligase like ACS (7;22;23). The family of ACS
includes short, medium, long and very long chain ACS (24). L-3HB could then be activated
by either medium chain ACS or acetoacetyl-CoA synthetases (AACS), although this has
proefschrift Soeters.indb 83 4-8-2008 9:13:21
84
not been investigated in human muscle (25). The possibility of D-3HB to be activated was
already suggested, although the catalyzing enzyme remained to be elucidated (8;9;23).
The second hypothetical pathway was activation of D-3-HB by SCOT. Indeed in muscle
homogenates of mice we found significant activation of D-3-HB via this pathway. L-3-
HB was less efficiently activated via this pathway. The robust activation of D-3HB in the
presence of succinyl-CoA strongly suggests that SCOT (activating AcAc in the presence
of succinyl-CoA) is involved in the synthesis of D-3-HB-carnitine in vivo in muscle (3;15).
This was father supported by the finding that the addition of succinyl-CoA to liver
homogenates did not result in a substantial amount of D-3HB-carnitine in accordance to
the absence of SCOT in adult liver (3;15). In addition, AACS has been shown to decrease
during fasting (26;27), at least in liver, making SCOT a better candidate to activate D-3HB
in skeletal muscle.
The second step is the transfer of carnitine to D-3HB-CoA. In mouse muscle and liver
homogenates, D-3HB-CoA was shown to be converted to D-3HB-carnitine. This reaction
was more efficient in liver compared to muscle homogenates. Finally, we tested if this
reaction is performed by CAT. Purified pigeon breast CAT was able to convert D-3HB-
CoA to D-3HB-carnitine, although the with low efficiency suggesting that other carnitine
acyltransferase might play a role (i.e. CPT2).
It is of interest where in the myocyte D-3HB is activated. Our results suggest that
most of the activation occurs by SCOT within the mitochondrion (3;15). Also subsequent
exchange of CoA for carnitine, possibly occurring via CAT or CPT2, is likely to be localized
in the mitochondrium in humans (28).
A remaining question is whether the increase in muscle D-3HB-carnitine serves a
purpose. First, D-3HB-carnitine can reflect true limitation of KB oxidation. It has been
suggested for short chain acyl-CoAs (i.e. acetyl-CoA) that the coupling to carnitine
provides an outlet to prevent mitochondrial accumulation (28;29). Finally, the increase
in muscle D-3HB-carnitine may serve different other purposes. In the first place, it was
shown that CoA levels in the rat heart, perfused with AcAc only, increase when carnitine
is co-administered (30). This suggests that the coupling of D-3HB to carnitine may reduce
CoA trapping. It was also suggested that the transfer of activated acylgroups to carnitine
provides in transport needs between compartments and a reservoir of energy (28).
Furthermore, ketosis can induce severe acidosis and it has been reported anecdotally
that fasting in disorders with systemic carnitine deficiency can induce severe keto-acidosis
suggesting that activation of D-3HB and subsequent transfer to carnitine may prevent
keto-acidosis during fasting (31).
proefschrift Soeters.indb 84 4-8-2008 9:13:22
D-3-hydroxybutyrylcarnitine: an alternative ketone body pathw
ay
85
Chapter 5
In conclusion we show in this study that D-3HB can be activated in muscle to its CoA
ester presumably by mitochondrial SCOT. After activation, CAT catalyzes the transfer of
CoA and carnitine thereby yielding D-3HB-carnitine. Hereby, D-3HB escapes immediate
mitochondrial oxidation. Furthermore we have shown that muscle D-3HB-carnitine
correlates with the plasma turnover of D-3HB in lean healthy men during ketosis induced
by 38 h of fasting. Since this has not been explored in detail before, our data provide a
newly identified alternative pathway of ketone body metabolism. The purpose of D-3HB-
carnitine synthesis and its fate remain to be elucidated. Besides an inability to oxidize all
formed D-3HB during ketosis, storage for intercompartmental transport, prevention of
keto-acidosis or decreased CoA trapping may all occur.
Reference List 1. Owen OE, Morgan AP, Kemp HG, Sullivan JM, Herrera MG, Cahill GF, Jr. Brain metabolism
during fasting. J Clin Invest 1967; 46(10):1589-1595.
2. Laffel L. Ketone bodies: a review of physiology, pathophysiology and application of monitoring to diabetes. Diabetes Metab Res Rev 1999; 15(6):412-426.
3. Fukao T, Lopaschuk GD, Mitchell GA. Pathways and control of ketone body metabolism: on the fringe of lipid biochemistry. Prostaglandins, Leukotrienes and Essential Fatty Acids 2004; 70(3):243-251.
4. Halestrap A, Meredith D. The SLC16 gene familyÇöfrom monocarboxylate transporters (MCTs) to aromatic amino acid transporters and beyond. Pfl++gers Archiv European Journal of Physiology 2004; 447(5):619-628.
5. An J, Muoio DM, Shiota M et al. Hepatic expression of malonyl-CoA decarboxylase reverses muscle, liver and whole-animal insulin resistance. Nat Med 2004; 10(3):268-274.
6. Hack A, Busch V, Pascher B et al. Monitoring of ketogenic diet for carnitine metabolites by subcutaneous microdialysis. Pediatr Res 2006; 60(1):93-96.
7. Swiatek KR, Dombrowski GJ, Chao KL, Chao HL. Metabolism of - and -3-hydroxybutyrate by rat liver during development. Biochemical Medicine 1981; 25(2):160-167.
8. Hildebrandt L, Spennetta T, Ackerman R, Elson C, Shrago E. Acyl CoA and Lipid Synthesis from Ketone Bodies by the Extramitochondrial Fraction of Hepatoma Tissue. Biochemical and Biophysical Research Communications 1996; 225(1):307-312.
9. Hildebrandt LA, Spennetta T, Elson C, Shrago E. Utilization and preferred metabolic pathway of ketone bodies for lipid synthesis by isolated rat hepatoma cells. Am J Physiol Cell Physiol 1995; 269(1):C22-C27.
10. Molven A, Matre GE, Duran M et al. Familial Hyperinsulinemic Hypoglycemia Caused by a Defect in the SCHAD Enzyme of Mitochondrial Fatty Acid Oxidation. Diabetes 2004; 53(1):221-227.
11. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 2003; 26(90001):5S-20.
12. Soeters MR, Sauerwein HP, Groener JE et al. Gender-Related Differences in the Metabolic Response to Fasting. J Clin Endocrinol Metab 2007; 92(9):3646-3652.
13. van Vlies N, Tian L, Overmars H et al. Characterization of carnitine and fatty acid metabolism in the long-chain acyl-CoA dehydrogenase-deficient mouse. Biochem J 2005; 387(1):185-193.
14. Minkler PE, Kerner J, Ingalls ST, Hoppel CL. Novel isolation procedure for short-, medium-, and long-chain acyl-coenzyme A esters from tissue. Analytical Biochemistry 2008; 376(2):275-276.
proefschrift Soeters.indb 85 4-8-2008 9:13:22
86
15. Fukao T, Mitchell GA, Song XQ et al. Succinyl-CoA:3-Ketoacid CoA Transferase (SCOT): Cloning of the Human SCOT Gene, Tertiary Structural Modeling of the Human SCOT Monomer, and Characterization of Three Pathogenic Mutations. Genomics 2000; 68(2):144-151.
16. Stannard SR, Thompson MW, Fairbairn K, Huard B, Sachinwalla T, Thompson CH. Fasting for 72 h increases intramyocellular lipid content in nondiabetic, physically fit men. Am J Physiol Endocrinol Metab 2002; 283(6):E1185-E1191.
17. Moller L, Jorgensen HS, Jensen FT, Jorgensen JO. Fasting in healthy subjects is associated with intrahepatic accumulation of lipids as assessed by 1H-magnetic resonance spectroscopy. Clin Sci (Lond) 2007.
18. Jensen MD, Ekberg K, Landau BR. Lipid metabolism during fasting. Am J Physiol Endocrinol Metab 2001; 281(4):E789-E793.
19. Scofield RF, Brady PS, Schumann WC et al. On the lack of formation of -(+)-3-hydroxybutyrate by liver. Archives of Biochemistry and Biophysics 1982; 214(1):268-272.
20. Reed WD, Ozand PT. Enzymes of -(+)-3-hydroxybutyrate metabolism in the rat. Archives of Biochemistry and Biophysics 1980; 205(1):94-103.
21. Tsai YC, Liao TH, Lee JA. Identification of -3-hydroxybutyrate as an original ketone body in rat serum by column-switching high-performance liquid chromatography and fluorescence derivatization. Analytical Biochemistry 2003; 319(1):34-41.
22. Lincoln BC, Rosiers CD, Brunengraber H. Metabolism of S-3-hydroxybutyrate in the perfused rat liver. Archives of Biochemistry and Biophysics 1987; 259(1):149-156.
23. Lehninger AL, Greville GD. The enzymic oxidation of d- and l-[beta]-hydroxybutyrate. Biochimica et Biophysica Acta 1953; 12(1-2):188-202.
24. Watkins PA. Fatty acid activation. Progress in Lipid Research 1997; 36(1):55-83.
25. Yamasaki M, Hasegawa S, Suzuki H, Hidai K, Saitoh Y, Fukui T. Acetoacetyl-CoA synthetase gene is abundant in rat adipose, and related with fatty acid synthesis in mature adipocytes. Biochemical and Biophysical Research Communications 2005; 335(1):215-219.
26. Bergstrom JD, Robbins KA, Edmond J. Acetoacetyl-coenzyme A synthetase activity in rat liver cytosol: A regulated enzyme in lipogenesis. Biochemical and Biophysical Research Communications 1982; 106(3):856-862.
27. Endemann G, Goetz PG, Edmond J, Brunengraber H. Lipogenesis from ketone bodies in the isolated perfused rat liver. Evidence for the cytosolic activation of acetoacetate. J Biol Chem 1982; 257(7):3434-3440.
28. Ramsay RR, Gandour RD, van der Leij FR. Molecular enzymology of carnitine transfer and transport. Biochim Biophys Acta 2001; 1546(1):21-43.
29. Lopaschuk GD, Belke DD, Gamble J, Toshiyuki I, Schonekess BO. Regulation of fatty acid oxidation in the mammalian heart in health and disease. Biochimica et Biophysica Acta (BBA) - Lipids and Lipid Metabolism 1994; 1213(3):263-276.
30. Russell RR, III, Taegtmeyer H. Coenzyme A sequestration in rat hearts oxidizing ketone bodies. J Clin Invest 1992; 89(3):968-973.
31. Boudin G, Mikol J, Guillard A, Engel AG. Fatal systemic carnitine deficiency with lipid storage in skeletal muscle, heart, liver and kidney. Journal of the Neurological Sciences 1976; 30(2-3):313-325.
proefschrift Soeters.indb 86 4-8-2008 9:13:22
proefschrift Soeters.indb 87 4-8-2008 9:13:22
proefschrift Soeters.indb 88 4-8-2008 9:13:22
Equal insulin sensitivity on inhibition of ketogenesis during short term-fasting in lean and obese humans6Maarten R. Soeters1, Hans P. Sauerwein1, Linda
Faas1, Martijn Smeenge1, Marinus Duran2,
Ronald J. Wanders2, An F.C. Ruiter3, Mariëtte
T. Ackermans3, Eric Fliers1, Sander M. Houten2,
and Mireille J. Serlie1
1Department of Endocrinology and Metabolism2 Department of Clinical Chemistry, Laboratory
Genetic Metabolic Diseases3 Department of Clinical Chemistry, Laboratory
of Endocrinology
Academic Medical Center, University of
Amsterdam, Amsterdam, the Netherlands
Submitted
proefschrift Soeters.indb 89 4-8-2008 9:13:22
Abstract
Aims/hypothesis: The ketone bodies D-3-hydroxybutyrate and acetoacetate play a
role in starvation and have been associated with insulin resistance: the dose-response
relationship between insulin and ketone bodies was demonstrated to be shifted to the
right in type 2 diabetes mellitus patients. In contrast ketone body levels have also been
reported to be decreased in obesity. To clarify this paradox, we investigated the metabolic
adaptation to fasting with respect to glucose and ketone body metabolism in lean and
obese men without non insulin dependent diabetes mellitus. We hypothesized ketone
body production to be equal under equal plasma insulin levels thereby reflecting absence
of insulin resistance on ketogenesis.
Methods: After 38 hours of fasting, glucose and total ketone body fluxes were measured
using stable isotopes in the basal state and during a two step pancreatic clamp.
Results: In the basal state, ketone body concentrations and fluxes were higher in lean
compared with obese men. During the pancreatic clamp, no differences were found in
ketone body fluxes during similar plasma insulin levels while peripheral glucose uptake
was lower in obese men.
Conclusions/interpretation: Obese subjects are resistant to insulin’s effect on
stimulation glucose of glucose uptake, but not its inhibiting effect on ketogenesis,
implying differential insulin sensitivity of intermediary metabolism in obesity.
proefschrift Soeters.indb 90 4-8-2008 9:13:22
Inhibition of ketogenesis in lean and obese humans
91
Chapter 6
Introduction
Starvation initiates an integrated metabolic response to prevent hypoglycemia and energy
depletion (1). The ketone bodies (KBs) D-3-hydroxybutyrate (D-3HB) and acetoacetate
(AcAc) play an important role in this adaptation: starvation induces an increase in plasma
KB concentration and turnover serving as an important fuel for the central nervous
system (2).
Ketogenesis occurs primarily in the liver. During fasting lipolysis rates from white adipose
tissue are increased, resulting in release of non esterified fatty acids (NEFA) into plasma
(3). These fatty acids are degraded through beta-oxidation within liver mitochondria,
resulting in the production of acetyl-CoA. Acetyl-CoA is either incorporated into the
tricarboxylic acid cycle or channelled into the ketogenesis pathway (4). The transport
of newly synthesized KBs from the hepatocytes to the circulation and the subsequent
uptake in extrahepatic tissues occurs via diffusion and the monocarboxylate transporters
1, 2 and 4 (5;6). Ketogenesis is strongly suppressed by insulin and is stimulated in states
of insulin deficiency and glucagon excess (4;6).
In addition to their physiological role in fasting, KBs have been associated with insulin
resistance: the dose-response relationship between circulating insulin and total KB (TKB)
concentration was demonstrated to be shifted to the right in type 2 diabetes mellitus
(DM) patients. This was interpreted as evidence of insulin resistance with respect to
ketogenesis (7). However, it has been shown that obesity is associated with lower plasma
D-3HB levels and a lower D-3HB oxidation compared with lean individuals (8-10). These
observations question the relevance of the reported association of KBs with insulin
resistance (7).
In this study we investigated the relation between total KB turnover and glucose
metabolism in 12 lean healthy and 12 obese non-diabetic men after short-term fasting (38
h), using stable isotopes and the pancreatic clamp technique. We hypothesized total KB
turnover rates (TKB Ra) not to be different between lean and obese subjects under equal
plasma insulin levels, indicating that insulin resistance has no effect on ketogenesis.
Subjects and methods
SubjectsTwelve healthy lean and 12 healthy obese male subjects were recruited via advertisements
in local magazines. Criteria for inclusion were 1) absence of a family history of diabetes
proefschrift Soeters.indb 91 4-8-2008 9:13:22
92
in the lean group but not in the obese group; 2) age 18–60 yr; 3) BMI 20-25 kg/m2 in
the lean group, BMI > 30 kg/m2 in the obese group; 4) normal oral glucose tolerance
test (OGTT) according to American Diabetes Association-criteria (11) in the lean group
whereas in the obese group glucose intolerance (but not diabetes) was accepted; 5)
no excessive sport activities, i.e. < 3 times per week; and 6) no medication. Subjects
were in self reported good health, confirmed by medical history, routine laboratory and
physical examination. Written informed consent was obtained from all subjects after
explanation of purposes, nature, and potential risks of the study. The study was approved
by the Medical Ethical Committee of the Academic Medical Center of the University of
Amsterdam.
Experimental protocolFor three days before the fasting period, all volunteers consumed a weight-maintaining
diet containing at least 250 g of carbohydrates per day. Then, the subjects started fasting
at 2000 h two days before the study day until the end of the study.
Volunteers were admitted to the metabolic unit of the Academic Medical Center of
the University of Amsterdam at 0730 h. Subjects were studied in the supine position and
were allowed to drink water only. A catheter was inserted into an antecubital vein for
infusion of stable isotope tracers, insulin, somatostatin, glucagon and glucose. Another
catheter was inserted retrogradely into a contralateral hand vein and kept in a thermo-
regulated (60ºC) plexiglas box for sampling of arterialized venous blood.
In all studies, saline was infused as NaCl 0.9% at a rate of 50 mL/h to keep the
catheters patent. [6,6-2H2]glucose (>99% enriched; Cambridge Isotopes, Andover, USA)
was used to study glucose kinetics. D[2,4-13C2]-3HB (>99% enriched; Cambridge Isotopes,
Andover, USA) was used to study KB turnover.
At T = 0 h (0800 h), blood samples were drawn for determination of background
enrichments and a primed continuous infusion the isotopes was started: [6,6-2H2]glucose
(prime, 8.8 μmol/kg; continuous, 0.11 μmol/kg·min) and D[2,4-13C2]-3HB (prime, 1.0
μmol/kg; continuous, 0.10 μmol/kg·min) and continued until the end of the study. After
an equilibration period of two hours (38 h of fasting), 3 blood samples were drawn for
glucose and KB enrichments and concentrations and 1 for glucoregulatory hormones
and NEFA. Thereafter (T = 2.5 h) a two-step pancreatic clamp was started with low dose
insulin infusions to suppress ketogenesis: step 1 included an infusion of insulin at a rate
of 4.5mU/m2·min (Actrapid 100 IU/ml; Novo Nordisk Farma B.V., Alphen aan den Rijn,
the Netherlands). Glucose 5% was started when necessary to maintain a plasma glucose
level of 5 mmol/L. [6,6-2H2]glucose was added to the 5% glucose solution to achieve
proefschrift Soeters.indb 92 4-8-2008 9:13:22
Inhibition of ketogenesis in lean and obese humans
93
Chapter 6
glucose enrichments of 1% to approximate the values for enrichment reached in plasma
and thereby minimizing changes in isotopic enrichment due to changes in the infusion
rate of exogenous glucose. At the same time (T=2.5 h) infusion of somatostatin (250
μg/h; Somatostatine-ucb; UCB Pharma, Breda, the Netherlands) was started to suppress
endogenous insulin and glucagon secretion, as well as a glucagon infusion (1 ng/kg·min);
Glucagen; Novo Nordisk, Alphen aan den Rijn, the Netherlands) to replace endogenous
glucagon concentrations. Somatostatin and glucagon infusions ran throughout step 1
and 2 of the clamp.
Plasma glucose levels were measured every 5 minutes at the bedside. After 2 h (T =
4.5 h), 5 blood samples were drawn at 5 minute intervals for determination of glucose
and D-3HB enrichments and concentrations. Another blood sample was drawn for
determination of glucoregulatory hormones and NEFA. Hereafter insulin infusion was
increased to a rate of 7.5mU/m2·min (step 2). Plasma glucose levels were continued to
be measured every 5 minutes and the final 5 blood samples were drawn at 5 minute
intervals for determination of glucose and D-3HB enrichments and concentrations (T =
6.5 h).
Body composition and indirect calorimetryBody composition was measured with bioelectrical impedance analysis (Maltron BF 906,
Rayleigh, UK). Respiratory energy expenditure was measured continuously during the
final 20 min of both the basal state and step 2 of the clamp by indirect calorimetry using
a ventilated hood system (Sensormedics model 2900; Sensormedics, Anaheim, USA).
Glucose, KB and NEFA measurementsPlasma glucose concentrations were measured with the glucose oxidase method using a
Beckman Glucose Analyzer 2 (Beckman, Palo Alto, USA, intra-assay variation 2-3%). KB
samples were drawn in chilled sodium-fluoride tubes and directly deproteinized with ice
cold perchloric acid 6% (1:1). AcAc and D-3HB were measured spectrophotometrically
using D-3-hydroxy-butyrate dehydrogenase (COBAS-FARA centrifugal analyzer, Roche
Diagnostics, Almere, the Netherlands). Plasma NEFA concentrations were determined
with an enzymatic colorimetric method (NEFA-C test kit, Wako Chemicals GmbH, Neuss,
Germany): intra-assay variation 1%; inter-assay variation: 4-15%; detection limit: 0.02
mmol/L.
[6,6-2H2]glucose enrichment was measured as described earlier (12). After
centrifugation of the acidified KB samples, the supernatants were neutralized with
KOH and HClO4. Following centrifugation the supernatants were acidified with HCl and
proefschrift Soeters.indb 93 4-8-2008 9:13:22
94
extracted twice with ethyl acetate. The combined extracts were dried under nitrogen at
room temperature. D-3HB and AcAc were converted to their t-butyldimethylsilyl derivatives
with MTBSTFA and pyridine. The sample was injected into an Agilent 6890/5973 MSD
gaschromatograph/mass spectrometer system (Agilent Technologies, Palo Alto, CA).
Separation was achieved on a Varian (Middelburg, The Netherlands) CP-SIL 19CB column
(30m x 0.25mm x 0.15μm). Selected ion monitoring (EI), data acquisition and quantitative
calculations were performed using the Agilent Chemstation software. Ions were monitored
at m/z 275 for unlabeled D-3HB, m/z 277 for D[2,4-13C2]-3HB, m/z 273 for unlabeled AcAc
and m/z 275 for [2,4-13C2]-AcAc. The enrichments in D-3HB and AcAc were determined
from standard curves of known enrichments injected in the same run.
Glucoregulatory hormonesInsulin was measured with a chemiluminescent immunometric assay (Immulite 2000
system, Diagnostic Products Corporation, Los Angeles, USA), intra-assay variation: 3-6%,
inter-assay variation: 4-6%, detection limit: 15 pmol/L. Basal state plasma insulin levels
were measured with an ultra sensitive human insulin RIA kit (Linco HI-11K, St. Charles,
Missouri, USA; Detection limit 1.5 pmol/L) in order to measure plasma insulin levels
below the detection limit of the chemiluminescent immunometric assay. Glucagon was
determined with the Linco 125
I radioimmunoassay (St. Charles, USA). Intra-assay variation
at 71 ng/L 10%, at 147 ng/L 9%; inter-assay variation at 84 ng/L 5%, at 192 ng/L 7%;
detection limit 15 ng/L. Cortisol, epinephrine and norepinephrine were determined as
described previously (12).
Calculations and statisticsResting energy expenditure (REE) and glucose and fat oxidation rates were calculated
from O2 consumption and CO2 production as reported previously (13). Endogenous
glucose production (EGP) and peripheral glucose uptake (rate of disappearance/Rd)
were calculated using the modified forms of the Steele Equations as described previously
(14;15). EGP and Rd were expressed as μmol/kg·min. TKB production (rate of appearance/
Ra) was calculated by using formulas for steady state as previously described (16;17):
proefschrift Soeters.indb 94 4-8-2008 9:13:22
Inhibition of ketogenesis in lean and obese humans
95
Chapter 6
Where F is the infusion rate (μmol/kg·min), [A] and [B] are the plasma concentrations
of AcAc and D-3HB respectively and TTR A and TTR B are the tracer tracee ratios of
AcAc and D-3BH respectively. TKB Ra was expressed as μmol/kg·min and as μmol/kcal as
proposed earlier for lipolysis (18). The KB ratio was calculated as [D-3HB]/[AcAc] (6). TKB
metabolic clearance rate (MCR) was calculated as TKB Ra/[TKB].
All data were analyzed with non parametric tests. Comparisons between groups (T =
2, 4.5 and 6.5 h) were performed using the Mann-Whitney U test. Comparisons within
groups (T = 2, 4.5 and 6.5 h) were performed with the Wilcoxon Signed Rank test.
Correlations were expressed as Spearman’s rank correlation coefficient (ρ). The SPSS
statistical software program version 14.0 (SPSS Inc, Chicago, IL) was used for statistical
analysis. Data are presented as median [minimum -maximum].
Results
Anthropometric characteristicsAnthropometric data are presented in Table 1. Lean men were younger compared to
obese men. As expected from the inclusion criteria there were significant differences
in weight, BMI, body fat mass and lean body mass (LBM). Fasting glucose levels were
not different between both groups, but glucose concentrations after the OGTT were
significantly higher in the obese subjects. Data on routine laboratory examinations prior
to inclusion are presented in Table 1.
Table 1 Subject characteristics
Non-obese (n=12) Obese (n=12) P
Age (y) 24 [20-53] 44 [30-54] <0.01
Length (m) 1.80 [ 1.71-1.89] 1.80 [1.72-1.92] 0.84
Weight (kg) 73.8 [59.0-83.5] 110.8 [95.0-124.5] <0.01
BMI (kg/cm2) 22.4 [18.9-24.7] 34.2 [31.0-38.4] <0.01
Body fat mass (%) 14.6 [7.5-22.9] 35.0 [30.3-42.5] <0.01
Lean body mass (%) 85.4 [77.1-92.5] 65.0 [57.5-69.7] <0.01
Fasting plasma glucose(mmol/liter) 4.6 [4.3-5.4]a 5.1 [4.5-6.4] 0.12
OGTT plasma glucose(mmol/liter) 4.4 [2.8-5.1]a 5.6 [2.8-9.6] 0.045
Plasma HDL-cholesterol (mmol/liter) 1.16 [0.95-2.09] 1.12 [0.77-1.48] 0.069
Plasma triacylglycerols (mmol/liter) 0.82 [0.20-2.65] 1.20 [0.93-3.31] 0.049
ALT (U/liter) 20 [8-46] 34 [17-100] 0.049
Data are expressed as median [min - max], aN = 11. ALT, alanine aminotransferase; OGGT, oral glucose tolerance test.
proefschrift Soeters.indb 95 4-8-2008 9:13:22
96
Resting energy expenditure, glucose, KB kinetics and NEFATotal REE was higher in obese subjects compared to lean subjects in the basal state:
2086 [1623 - 2200] kcal/day vs. 1797 [1428 - 2067] kcal/day respectively, P < 0.01; as
well as at the end of step 2 of the clamp : 1952 [1226 - 2107] vs. 1682 [1293 - 1958]
respectively, P < 0.01.
Plasma glucose levels were significantly higher in the obese subjects compared to the
lean subjects in the basal state and step 1 of the clamp (Table 2). During step 2, plasma
glucose levels were not different. EGP was significantly lower in obese subjects in the
basal state after 38 h of fasting (Figure 1). No change in EGP was observed between
step 1 and the basal state in obese subjects (P = 0.24), but a significant decrease was
observed for step 2 compared to step 1 (P = 0.015). EGP decreased in the lean subjects
during the clamp; step 1 vs. basal state and step 2 vs. step 1: P < 0.01 and P < 0.01
respectively. Rd of glucose was lower during all time points in obese men compared to
lean controls (Table 2).
Data on KB are presented in Table 2 and Figure 1. Plasma D-3HB was lower in obese
men in the basal state, but no differences were found during the clamp; plasma AcAc
tended to be lower in obese subjects in the basal state but no differences were observed
between groups during step 1 of the clamp. During step 2, plasma AcAc tended to be
higher in obese subjects. TKB Ra (in either μmol/kg·min or μmol/kcal) was significantly
lower in obese subjects in the basal state but during the clamp no differences were
observed (Figure 1). TKB Ra decreased within lean subjects during the clamp; step 1 vs.
Table 2 Glucose, ketone body and lipid metabolism measurements.
basal state clamp (step 1) clamp (step 2)
lean men(n = 12)
obese men(n = 12)
P lean men(n = 12)
obese men(n = 12)
P lean men(n = 12)
obese men(n = 12)
P
Glucose (mmol/liter) 4.1 [3.5 - 4.5] 5.2 [4.2 - 5.5] <0.01 4.8 [4.2 - 5.5] 6.3 [4.7 - 8.2] <0.01 5.0 [4.5 - 5.2] 5.5 [4.5 - 7.3] 0.119
Rd (μmol/kg•min) - - - 12.8 [10.3 - 22.7] 8.0 [4.6 - 11.7] <0.01 12.4 [10.3 - 18.6] 7.1 [3.6 - 10.2] <0.01
Rd (μmol/kg LBM•min) - - - 16.1 [12.3 - 25.6] 12.5 [7.0 - 17.3] 0.017 14.7 [11.9 - 22.7] 10.8 [5.2 - 15.0] <0.01
D-3HB (mmol/liter) 1.24 [0.30 - 2.03] 0.25 [0.09 - 1.40] <0.01 0.35 [0.02 - 1.00] 0.30 [0.14 - 0.90] 0.93 0.05 [0 - 0.40] 0.09 [0.03 - 0.60] 0.11
AcAc (mmol/liter) 0.28 [0.08 - 0.69] 0.16 [0.04 - 0.53] 0.069 0.11 [0.01 - 0.43] 0.15 [0.04 - 0.40] 0.62 0.04 [0 - 0.17] 0.08 [0.01 - 0.34] 0.073
TKBc (mmol/liter) 1.54 [0.38 - 2.57] 0.41 [0.16 - 1.93] <0.01 0.45 [0.03 - 1.43] 0.43 [0.18 - 1.30] 0.84 0.10 [0 - 0.54] 0.20 [0.04 - 0.94] 0.088
TKB Ra (μmol/kcal) 1.3 [0.8 - 2.5] 0.6 [0.3 - 1.2] <0.01 - - - 0.3 [0 - 0.7] 0.3 [0.1 - 1.6] 0.33
MCR TKB (ml/kg·min) 10.9 [5.5 - 26.6] 11.8 [5.1 - 216.7] 0.91 13.2 [6.9 - 55.1] 11.5 [6.8 - 15.0] 0.33 24.3 [0 - 329.3] 13.6 [8.3 - 25.3] 0.039
NEFA (mmol/liter) 1.04 [0.72 - 1.21] 0.75 [0.48 - 1.29] <0.01 0.39 [0.18 - 0.49] 0.52 [0.39 - 0.79] <0.01 0.22 [0.06 - 0.48] 0.31 [0.23 - 0.63] 0.012
Rd rate of disappearance; MCR metabolic clearance rate; TKBc total ketone body concentration. Data are presented as median [minimum - maximum].
proefschrift Soeters.indb 96 4-8-2008 9:13:22
Inhibition of ketogenesis in lean and obese humans
97
Chapter 6
basal state and step 2 vs. step 1: P < 0.01 and P < 0.01 respectively. No change in TKB
Ra was observed between step 1 and the basal state in obese subjects (P = 0.48), but a
significant decrease was observed for step 2 compared to step 1 (P < 0.01). When TKB
Ra was expressed as μmol/kg LBM ·min comparable results were obtained (data not
shown).
Table 2 Glucose, ketone body and lipid metabolism measurements.
basal state clamp (step 1) clamp (step 2)
lean men(n = 12)
obese men(n = 12)
P lean men(n = 12)
obese men(n = 12)
P lean men(n = 12)
obese men(n = 12)
P
Glucose (mmol/liter) 4.1 [3.5 - 4.5] 5.2 [4.2 - 5.5] <0.01 4.8 [4.2 - 5.5] 6.3 [4.7 - 8.2] <0.01 5.0 [4.5 - 5.2] 5.5 [4.5 - 7.3] 0.119
Rd (μmol/kg•min) - - - 12.8 [10.3 - 22.7] 8.0 [4.6 - 11.7] <0.01 12.4 [10.3 - 18.6] 7.1 [3.6 - 10.2] <0.01
Rd (μmol/kg LBM•min) - - - 16.1 [12.3 - 25.6] 12.5 [7.0 - 17.3] 0.017 14.7 [11.9 - 22.7] 10.8 [5.2 - 15.0] <0.01
D-3HB (mmol/liter) 1.24 [0.30 - 2.03] 0.25 [0.09 - 1.40] <0.01 0.35 [0.02 - 1.00] 0.30 [0.14 - 0.90] 0.93 0.05 [0 - 0.40] 0.09 [0.03 - 0.60] 0.11
AcAc (mmol/liter) 0.28 [0.08 - 0.69] 0.16 [0.04 - 0.53] 0.069 0.11 [0.01 - 0.43] 0.15 [0.04 - 0.40] 0.62 0.04 [0 - 0.17] 0.08 [0.01 - 0.34] 0.073
TKBc (mmol/liter) 1.54 [0.38 - 2.57] 0.41 [0.16 - 1.93] <0.01 0.45 [0.03 - 1.43] 0.43 [0.18 - 1.30] 0.84 0.10 [0 - 0.54] 0.20 [0.04 - 0.94] 0.088
TKB Ra (μmol/kcal) 1.3 [0.8 - 2.5] 0.6 [0.3 - 1.2] <0.01 - - - 0.3 [0 - 0.7] 0.3 [0.1 - 1.6] 0.33
MCR TKB (ml/kg·min) 10.9 [5.5 - 26.6] 11.8 [5.1 - 216.7] 0.91 13.2 [6.9 - 55.1] 11.5 [6.8 - 15.0] 0.33 24.3 [0 - 329.3] 13.6 [8.3 - 25.3] 0.039
NEFA (mmol/liter) 1.04 [0.72 - 1.21] 0.75 [0.48 - 1.29] <0.01 0.39 [0.18 - 0.49] 0.52 [0.39 - 0.79] <0.01 0.22 [0.06 - 0.48] 0.31 [0.23 - 0.63] 0.012
Rd rate of disappearance; MCR metabolic clearance rate; TKBc total ketone body concentration. Data are presented as median [minimum - maximum].
FIGURE 1 Panel A: Differences between obese (open box plots) and lean (grey box plots) subjects in endogenous glucose production (EGP, μmol/kg•min) during the basal state and during step 1 and step 2 of the clamp; ap < 0.01, bp <0.01 and cp = 0.01 respectively.Panel B: Differences between obese (open box plots) and lean (grey box plots) subjects in total ketone body turnover (TKB Ra, μmol/kg•min) during the basal state and during step 1 and step 2 of the clamp; ap < 0.01, bp = 0.64 and cp = 0.77 respectively.
proefschrift Soeters.indb 97 4-8-2008 9:13:23
98
The MCRTKB showed no differences in the basal state and step 1 of the clamp (Table
2). During step 2, obese subjects had lower MCRTKB compared to lean subjects. The KB
ratio in the basal state was lower in obese subjects compared to lean subjects: 2.1 [1.3
– 3.8] vs. 3.8 [1.9 - 7.0] respectively, P < 0.01. The KB ratio’s during the clamp were
not different between obese and lean subjects (step 1: 2.1 [1.1 – 3.8] vs. 2.2 [1.0 - 4.3]
respectively, P = 0.54; step 2: 1.7 [0.6 – 4.0] vs. 1.5 [0 - 3.5] respectively, P = 0.31).
Obese men had lower plasma NEFA levels in the basal state, but higher plasma NEFA
levels during the clamp compared to lean men (Table 2). In the basal state, plasma NEFA
correlated with TKB Ra in lean men; a trend towards a correlation was found in obese
men (Figure 2).
Table 3 Glucoregulatory hormones
basal state clamp (step 1) clamp (step 2)
lean men(n = 12)
obese men(n = 12)
P lean men(n = 12)
obese men(n = 12)
P lean men(n = 12)
obese men(n = 12)
P
Insulin (pmol/liter) 8 [8-25] 65 [8-102] <0.01 33 [24-40] 40 [15-50] 0.06 49 [26-69] 58 [41-79] 0.31
Glucagon (ng/liter) 75 [66-123] 86 [66-107]a 0.26 73 [58-84] 91 [60-125] <0.01 67 [48-88] 90 [53-122]a <0.01
Cortisol (nmol/liter) 277 [138 - 422] 339 [98 - 726] 0.51 206 [89 - 505] 191 [65 - 773] 1.0 182 [125 - 494] 286 [151 - 629] 0.033
Epinephrine (nmol/liter) 0.28 [0.06 - 0.45] 0.11 [0.05 - 0.58] 0.056 0.095 [0.05 - 0.31] 0.05 [0.05 - 0.92] 0.16 0.105 [0.05 - 0.37] 0.08 [0.05 - 1.15] 0.63
Norepinephrine (nmol/liter) 0.70 [0.19 - 1.15] 1.11 [0.41 - 1.41] <0.01 0.56 [0.1 - 1.07] 1.04 [0.64 - 1.38] <0.01 0.46 [0.06 - 1.48] 0.88 [0.31 - 1.35] 0.06
Data are presented as median [minimum - maximum]. aN = 11.
FIGURE 2 Panel A: Correlations of plasma NEFA levels with TKB Ra in the basal state for obese subjects ( ):ρ = 0.55, and p = 0,063; and lean subjects (l):ρ = 0.76, and p < 0.01.Panel B: Correlations of plasma insulin levels with TKB Ra in the basal state for obese subjects ( ):ρ = -0.78, and p < 0.01; and lean subjects (l):ρ = -0.24, and p = 0.46.
proefschrift Soeters.indb 98 4-8-2008 9:13:23
Inhibition of ketogenesis in lean and obese humans
99
Chapter 6
Glucoregulatory hormonesInsulin levels were higher in obese subjects compared to lean subjects in the basal state
(Table 3). During step 1, a trend towards higher insulin levels in obese compared to lean
subjects was found, however there was no difference during step 2 (Table 3). In the basal
state there was a significant negative correlation between plasma insulin levels and TKB
Ra in obese but not in lean men (Figure 2).
Plasma glucagon levels were not different in the basal state, but were significantly
higher in obese subjects during the clamp. Cortisol levels were significantly higher in
obese subjects during step 2 of the clamp but not at other time points (Table 3). The
epinephrine concentration tended to be lower in obese men in the basal state; no
differences were found during step 1 and 2 of the clamp. Norepinephrine levels were
higher in obese subjects in the basal state and step 1 of the clamp and tended to be
higher during step 2 of the clamp.
Discussion
In patients with type 2 DM, a right shifted dose response curve for insulin versus ketogenesis
has been reported, suggesting insulin resistance on ketogenesis (7). The present study
was designed to investigate ketone body metabolism in obese insulin resistant subjects
after 38 h of fasting in relation to glucose metabolism. We found that, after short-term
fasting, rates of ketogenesis are higher in lean subjects but this difference is no longer
present after reaching similar plasma insulin levels. This implies that there is no insulin
resistance on ketogenesis in obese subjects during short-term fasting
We confirmed earlier data on lower EGP in obese subjects in the basal state after 38
h of fasting due to higher plasma insulin levels (9). We did not express EGP as μmol/kg
Table 3 Glucoregulatory hormones
basal state clamp (step 1) clamp (step 2)
lean men(n = 12)
obese men(n = 12)
P lean men(n = 12)
obese men(n = 12)
P lean men(n = 12)
obese men(n = 12)
P
Insulin (pmol/liter) 8 [8-25] 65 [8-102] <0.01 33 [24-40] 40 [15-50] 0.06 49 [26-69] 58 [41-79] 0.31
Glucagon (ng/liter) 75 [66-123] 86 [66-107]a 0.26 73 [58-84] 91 [60-125] <0.01 67 [48-88] 90 [53-122]a <0.01
Cortisol (nmol/liter) 277 [138 - 422] 339 [98 - 726] 0.51 206 [89 - 505] 191 [65 - 773] 1.0 182 [125 - 494] 286 [151 - 629] 0.033
Epinephrine (nmol/liter) 0.28 [0.06 - 0.45] 0.11 [0.05 - 0.58] 0.056 0.095 [0.05 - 0.31] 0.05 [0.05 - 0.92] 0.16 0.105 [0.05 - 0.37] 0.08 [0.05 - 1.15] 0.63
Norepinephrine (nmol/liter) 0.70 [0.19 - 1.15] 1.11 [0.41 - 1.41] <0.01 0.56 [0.1 - 1.07] 1.04 [0.64 - 1.38] <0.01 0.46 [0.06 - 1.48] 0.88 [0.31 - 1.35] 0.06
Data are presented as median [minimum - maximum]. aN = 11.
proefschrift Soeters.indb 99 4-8-2008 9:13:23
100
LBM ·min since glucose utilization during short-term fasting by skeletal muscle is typically
low in contrast to glucose utilization by the central nervous system (19).
Plasma D-3HB and TKB levels as well as TKB Ra were higher in lean subjects compared
to obese subjects in the basal state as shown earlier (8-10). There are multiple ways to
explain this difference. First, higher insulin levels can inhibit the activity of 3-hydroxy-3-
methylglutaryl-CoA synthase (mHS) in the liver (4;6;20). Obesity-induced insulin resistance
is characterized by higher plasma NEFA levels that induce hepatic and peripheral insulin
resistance and subsequent hyperinsulinemia (21). Indeed plasma insulin levels correlated
negatively with TKB Ra in the obese subjects in the basal state, whereas the lack of this
correlation in lean subjects may be explained by the suppressed plasma insulin levels after
38 h of fasting. Secondly, it has been demonstrated that plasma glucose levels inhibit
ketogenesis independently from insulin (22). It is unlikely, however, that the higher
plasma glucose levels in the obese volunteers induced lower plasma KB levels, since the
suppressive effect of glucose on KBs was observed during profound hyperglycemia (i.e.
12 mmol/L) (22). Finally, and more importantly, the lower TKB Ra in obese subjects in
the basal state may be explained by lower NEFA levels for these are reported to increase
ketone body production independently of insulin (17). Besides decreasing hepatic
ketogenesis, insulin also inhibits lipolysis by inhibiting hormone sensitive lipase (HSL) in
adipose tissue (4;6). Indeed we found a convincing correlation of NEFA and TKB Ra in the
basal state in lean subjects and a trend towards a correlation in obese subjects.
Initially, MCRTKB did not differ between lean and obese men which emphasizes the
notion that the rate of utilization of KBs is proportional to their circulating levels (4).
However during step 2, MCRTKB was lower in obese subjects, explaining the trend
towards higher plasma TKB concentrations at that time point.
Our study does not support the data by Singh et al (7) who demonstrated a significant
right shift of the dose-response relationship between circulating insulin and TKB
concentrations in type 2 diabetes mellitus patients with DM2. This can be explained
by differences in study design since we studied obese insulin resistant subjects with
approximately four-fold higher insulin levels in the basal state, whereas the plasma insulin
levels in the patients of Singh were similar to lean controls. Furthermore our volunteers
had been fasting for 38 h opposed to the overnight fast in their study (7).
In this study we used a single isotope method to analyze TKB Ra (16). Earlier, the
use of a single isotope model in measuring TKB Ra was debated because of the rapid
interconversion of AcAc and D-3HB (25). Later, it was discussed by Beylot et al that
the single isotope method provided a reasonable estimate of TKB Ra even in the post-
absorptive state when interconversion between AcAc and D-3HB is relatively high (16;17).
proefschrift Soeters.indb 100 4-8-2008 9:13:23
Inhibition of ketogenesis in lean and obese humans
101
Chapter 6
Our volunteers had plasma KB levels well above post-absorptive levels indicating less
interconversion of AcAc and D-3HB thereby permitting a single isotope analysis (16).
Moreover during the clamp KB ratios were not different. On theoretical grounds our
results in the basal state could be overestimated in the lean group by 15%. However the
enormous difference in TKB Ra between the lean and obese subjects in the basal state
can not be due to such overestimation.
The studied groups were not matched for age. It has been shown that aging in
healthy men is associated with increased KB levels during prolonged fasting, but not in
the post-absorptive state (26). Therefore, our findings are not likely to be attributed to
age differences in our groups.
Plasma glucagon levels were not different in the basal state, but this does not preclude
glucagon from being a regulator of ketosis in lean subjects. Although the exact role of
glucagon in ketogenesis is unclear to date (17;27), Miles et al showed that glucagon
increases KB production when NEFA concentrations are increased in the setting of
isolated insulin deficiency (27). This may explain why the higher plasma glucagon levels
during the clamp did not result in higher TKB Ra in obese men compared with lean
men. Furthermore we witnessed a trend towards higher plasma epinephrine levels in
lean subjects in the basal state and higher plasma norepinephrine levels throughout
the day. Epinephrine increases ketogenesis via lipolysis (4;28), but norepinephrine may
induce ketogenesis directly (29). TKB Ra was not significantly different during the clamps,
suggesting that norepinephrine in the presence of higher insulin levels is not a major
drive of ketogenesis in this setting.
In conclusion, we show in this study that ketogenesis rates are equally sensitive for
insulin in obese and lean subjects. Meanwhile obese subjects display a lower peripheral
insulin sensitivity of insulin mediated glucose uptake supporting differential insulin hepatic
sensitivity of KB metabolism and glucose metabolism in obese non diabetic men.
Reference List 1. Cahill GF, Jr., Herrera MG, Morgan AP, et al (1966) Hormone-fuel interrelationships during
fasting. J Clin Invest. 45: 1751-1769
2. Owen OE, Morgan AP, Kemp HG, Sullivan JM, Herrera MG, Cahill GF, Jr. (1967) Brain metabolism during fasting. J Clin.Invest 46: 1589-1595
3. Klein S, Sakurai Y, Romijn JA, Carroll RM (1993) Progressive alterations in lipid and glucose metabolism during short-term fasting in young adult men. Am J Physiol Endocrinol Metab 265: E801-E806
proefschrift Soeters.indb 101 4-8-2008 9:13:23
102
4. Fukao T, Lopaschuk GD, Mitchell GA (2004) Pathways and control of ketone body metabolism: on the fringe of lipid biochemistry. Prostaglandins, Leukotrienes and Essential Fatty Acids 70: 243-251
5. Halestrap A, Meredith D (2004) The SLC16 gene familyGÇöfrom monocarboxylate transporters (MCTs) to aromatic amino acid transporters and beyond. Pflügers Archiv European Journal of Physiology 447: 619-628
6. Laffel L (1999) Ketone bodies: a review of physiology, pathophysiology and application of monitoring to diabetes. Diabetes Metab Res.Rev 15: 412-426
7. Singh BM, Krentz AJ, Nattrass M (1993) Insulin resistance in the regulation of lipolysis and ketone body metabolism in non-insulin dependent diabetes is apparent at very low insulin concentrations. Diabetes Research and Clinical Practice 20: 55-62
8. Vice E, Privette JD, Hickner RC, Barakat HA (2005) Ketone body metabolism in lean and obese women. Metabolism 54: 1542-1545
9. Bergman BC, Cornier MA, Horton TJ, Bessesen DH (2007) Effects of fasting on insulin action and glucose kinetics in lean and obese men and women. Am J Physiol Endocrinol Metab 293: E1103-E1111
10. Elia M, Stubbs RJ, Henry CJ (1999) Differences in fat, carbohydrate, and protein metabolism between lean and obese subjects undergoing total starvation. Obes.Res. 7: 597-604
11. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus (2003). Diabetes Care 26: 5S-20
12. Soeters MR, Sauerwein HP, Groener JE, et al (2007) Gender-Related Differences in the Metabolic Response to Fasting. J Clin Endocrinol Metab 92: 3646-3652
13. Frayn KN (1983) Calculation of substrate oxidation rates in vivo from gaseous exchange. J Appl Physiol 55: 628-634
14. Finegood DT, Bergman RN, Vranic M (1987) Estimation of endogenous glucose production during hyperinsulinemic-euglycemic glucose clamps. Comparison of unlabeled and labeled exogenous glucose infusates. Diabetes. 36: 914-924
15. Steele R (1959) Influences of glucose loading and of injected insulin on hepatic glucose output. Ann.N Y.Acad.Sci. 82:420-30.: 420-430
16. Beylot M, Beaufrere B, Normand S, Riou JP, Cohen R, Mornex R (1986) Determination of human ketone body kinetics using stable-isotope labelled tracers. Diabetologia 29: 90-96
17. Beylot M, Picard S, Chambrier C, et al (1991) Effect of physiological concentrations of insulin and glucagon on the relationship between nonesterified fatty acids availability and ketone body production in humans. Metabolism 40: 1138-1146
18. Koutsari C, Jensen MD (2006) Thematic review series: patient-oriented research. Free fatty acid metabolism in human obesity. J Lipid Res. 47: 1643-1650
19. Cahill J (1976) Starvation in man. Clinics in Endocrinology and Metabolism 5: 397-415
20. Hegardt FG (1999) Mitochondrial 3-hydroxy-3-methylglutaryl-CoA synthase: a control enzyme in ketogenesis. Biochem.J. 338: 569-582
21. Boden G, Shulman GI (2002) Free fatty acids in obesity and type 2 diabetes: defining their role in the development of insulin resistance and beta-cell dysfunction. Eur.J Clin Invest 32 Suppl 3: 14-23
22. Riou JP, Beylot M, Laville M, et al (1986) Antiketogenic effect of glucose per se in vivo in man and in vitro in isolated rat liver cells. Metabolism 35: 608-613
23. Beylot M, Khalfallah Y, Riou JP, Cohen R, Normand S, Mornex R (1986) Effects of ketone bodies on basal and insulin-stimulated glucose utilization in man. J.Clin.Endocrinol.Metab 63: 9-15
24. Miles JM, Haymond MW, Gerich JE (1981) Suppression of glucose production and stimulation of insulin secretion by physiological concentrations of ketone bodies in man. J Clin Endocrinol Metab 52: 34-37
25. Barton RN (1980) Isotopic study of ketone body kinetics: Invalidity of calculations based upon specific radioactivity of total ketone bodies. Metabolism 29: 392-394
proefschrift Soeters.indb 102 4-8-2008 9:13:23
Inhibition of ketogenesis in lean and obese humans
103
Chapter 6
26. London ED, Margolin RA, Duara R, et al (1986) Effects of fasting on ketone body concentrations in healthy men of different ages. J Gerontol. 41: 599-604
27. Miles JM, Haymond MW, Nissen SL, Gerich JE (1983) Effects of free fatty acid availability, glucagon excess, and insulin deficiency on ketone body production in postabsorptive man. J Clin Invest 71: 1554-1561
28. Beylot M (1996) Regulation of in vivo ketogenesis: role of free fatty acids and control by epinephrine, thyroid hormones, insulin and glucagon. Diabetes Metab 22: 299-304
29. Keller U, Lustenberger M, Muller-Brand J, Gerber PP, Stauffacher W (1989) Human ketone body production and utilization studied using tracer techniques: regulation by free fatty acids, insulin, catecholamines, and thyroid hormones. Diabetes Metab Rev 5: 285-298
proefschrift Soeters.indb 103 4-8-2008 9:13:23
proefschrift Soeters.indb 104 4-8-2008 9:13:23
Intermittent fasting differentially affects glucose, lipid and protein metabolism.7Maarten R. Soeters1, Nicolette M. Lammers,
Peter F. Dubbelhuis2, Mariëtte T. Ackermans3,
Cora F. Jonkers-Schuitema4, Eric Fliers1, Hans P.
Sauerwein1 and Mireille J. Serlie1
1Department of Endocrinology and Metabolism2Department of Medical Biochemistry3Department of Clinical Chemistry, Laboratory
of Endocrinology4Department of Dietetics
Academic Medical Center, University of
Amsterdam, Amsterdam, the Netherlands
Submitted
proefschrift Soeters.indb 105 4-8-2008 9:13:23
Abstract
Context: Intermittent fasting (IF) was shown to increase whole body insulin sensitivity
but it is uncertain whether IF selectively influences intermediary metabolism. Such
selectivity might be advantageous when adapting to periods of food abundance and
food shortage.
Objective: To assess effects of IF on intermediary metabolism and energy expenditure.
Main Outcome Measures and Design: Glucose, glycerol and valine fluxes were
measured after two weeks of IF and a standard diet (SD), before and during a two-
step hyperinsulinemic euglycemic clamp with assessment of energy expenditure and
phosphorylation of muscle protein kinase B (AKT/PKB), glycogen synthase kinase (GSK)
and mTOR. We hypothesized that IF selectively increases peripheral glucose uptake and
lowers proteolysis thereby protecting protein stores.
Results: There were no differences in body weight after IF vs. SD. Peripheral glucose
uptake during step 1, but not step 2, was significantly higher after IF while hepatic insulin
sensitivity did not differ. Insulin-mediated suppression of lipolysis tended to be increased
after IF. Proteolysis was significantly lower after IF, but only at higher insulin levels. IF
decreased resting energy expenditure. Finally, IF tended to increase phosphorylation of
AKT, and significantly increased phosphorylation of glycogen synthase kinase.
Conclusion: IF increases peripheral, but not hepatic insulin sensitivity and tends to
increase the anti-lipolytic effect of insulin. Proteolysis is lower under hyperinsulinemic
conditions. The decrease in REE after IF warns for increasing weight during IF when
caloric intake is not adjusted. Whether IF is beneficial in improving peripheral insulin
resistance in obese insulin resistant subjects remains elusive.
proefschrift Soeters.indb 106 4-8-2008 9:13:23
Intermittent fasting and glucose, lipid and protein m
etabolism
107
Chapter 7
Introduction
Intermittent fasting (IF) has been suggested to mimic cycles of feast and famine (food
abundance and food shortage) as may have been physiological in the late Palaeolithic
era (1). Accordingly, it has been postulated that humans have developed metabolic
pathways that oscillate with the cycles of feast-famine and physical activity-rest (2). Such
adaptive mechanisms were proposed to be part of a thrifty genotype, necessary for
survival during shortage of food, protecting muscle lipid and glycogen stores during
fasting while replenishing fuel stores during refeeding (2;3).
Previous human studies on the metabolic effects of IF are scarce and unequivocal
(1;4;5). However, the increase of whole body insulin sensitivity after two weeks of IF in
healthy volunteers as shown by Halberg et al is of interest since it may provide in a simple
tool to improve insulin sensitivity in subjects with insulin resistance (1).
IF consists of repetitive bouts of short-term fasting. The latter is characterized by
integrated alterations in intermediary metabolism that guarantee substrate availability
for energy production necessary for survival (6-9). Specifically, fasting induces a reduction
in peripheral insulin sensitivity via a decrease in phosphorylation of both protein kinase B
(AKT) and AS160 after short-term fasting (8). Hepatic insulin sensitivity may be affected
by fasting as well (10), but not all studies have confirmed this (11). To date, it is uncertain
whether IF influences hepatic insulin sensitivity (1).
Furthermore, short-term fasting increases proteolysis (12;13) and decreases protein
synthesis (14). The fall in protein synthesis and the rise in proteolysis during short-term
fasting occur through decreased activation of AKT and the downstream mammalian
target of rapamycine (mTOR) (15). It is unknown whether IF affects protein metabolism
via changes in activity of mTOR, at present.
In this study, we measured glucose, glycerol and valine fluxes after two weeks IF and
after two weeks of a standard isocaloric diet (SD) both in the basal state and during a
two-step hyperinsulinemic euglycemic clamp (stable isotope technique). Immunoblots
were performed of crucial signaling proteins in the insulin and mTOR signaling pathways
in muscle tissue samples obtained in the basal state and during the clamp. It was our
hypothesis that IF selectively increases peripheral insulin sensitivity of glucose uptake via
increased phosphorylation of AKT and AS160. Additionally we expected proteolysis to
be lower after IF thereby protecting protein stores.
proefschrift Soeters.indb 107 4-8-2008 9:13:23
108
Subjects and methods
SubjectsEight men were recruited via advertisements in local magazines. Subjects were in self-
reported good health, confirmed by medical history and physical examination. Criteria for
inclusion were 1) absence of a family history of diabetes; 2) age 18–35 yr; 3) Caucasian
race; 4) BMI 20-25 kg/m2; 5) normal oral glucose tolerance test according to the ADA
criteria (16); 6) normal routine blood examination; 7) no excessive sport activities, i.e.
< 3 times per week; and 8) no medication. Additionally, subjects who were not in the
habit of eating breakfast every day were excluded. At inclusion, indirect calorimetry and
dietary interviews were performed by a dietician to assess daily energy requirements.
Written informed consent was obtained from all subjects after explanation of purposes,
nature, and potential risks of the study. The study was approved by the Medical Ethical
Committee of the Academic Medical Center of the University of Amsterdam.
Experimental protocolSubjects were studied twice in balanced assignment (cross over-design) to two weeks of
intermittent fasting or two weeks of a standard diet. Study days were separated by at
least four weeks to minimize influences of the previous diet. Subjects were fasting from
2000 h the evening before the study days. They were allowed to drink water only.
After admission at the Metabolic Unit at 0730 h, a catheter was inserted into an
antecubital vein for infusion of stable isotope tracers, insulin and glucose. Another
catheter was inserted retrogradely into a contralateral hand vein and kept in a thermo-
regulated (60ºC) plexiglas box for sampling of arterialized venous blood. Saline was
infused as NaCl 0.9% at a rate of 50 mL/h to keep the catheters patent. [6,6-2H2]
glucose, [1,1,2,3,3-2H5]glycerol and L-[1-13C]valine were used as tracers (>99% enriched;
Cambridge Isotopes, Andover, USA) to study glucose kinetics, lipolysis (total triglyceride
hydrolysis), and proteolysis respectively.
At T = 0 h (0800 h), blood samples were drawn for determination of background
enrichments. Then a primed continuous infusion of isotopes was started: [6,6-2H2]
glucose (prime, 8.8 μmol/kg; continuous, 0.11 μmol/kg·min), [1,1,2,3,3-2H5]glycerol
(prime, 1.6 μmol/kg; continuous, 0.11 μmol/kg·min) and L-[1-13C]valine (prime, 13.7
μmol/kg; continuous, 0.15 μmol/kg·min) and continued until the end of the study. After
an equilibration period of two hours (14 h of fasting), 3 blood samples were drawn for
isotope enrichments and 1 for glucoregulatory hormones and FFA. Thereafter (T = 3.0 h)
a two-step hyperinsulinemic euglycemic clamp was started: step 1 included an infusion
proefschrift Soeters.indb 108 4-8-2008 9:13:24
Intermittent fasting and glucose, lipid and protein m
etabolism
109
Chapter 7
of insulin at a rate of 10 mU/m2·min (Actrapid 100 IU/ml; Novo Nordisk Farma B.V.,
Alphen aan den Rijn, the Netherlands) to assess hepatic insulin sensitivity. Glucose 10%
was started to maintain a plasma glucose level of 5 mmol/L. [6,6-2H2]glucose was added
to the 10% glucose solution to achieve glucose enrichments of 1% to approximate the
values for enrichment reached in plasma and thereby minimizing changes in isotopic
enrichment due to changes in the infusion rate of exogenous glucose (17).
Plasma glucose levels were measured every 5 minutes at the bedside. After 2
h (T = 5 h), 5 blood samples were drawn at 5 minute intervals for determination of
glucose concentrations and isotopic enrichments. Another blood sample was drawn
for determination of glucoregulatory hormones and FFA. Hereafter insulin infusion was
increased to a rate of 40 mU/m2·min (step 2) to assess peripheral insulin sensitivity. After
another 2 h (T = 7 h), blood sampling was repeated.
Intermittent fasting and standard dietThe two weeks of IF were equal to the study of Halberg et al (1); subjects were fasting
every second day for 20 h (Figure 1). Fasting started at 2200 and ended at 1800 h the
following day. In total, subjects fasted seven times. During SD, subjects were not allowed
to skip meals. The day before the study days, subjects were not fasting, but consumed
their diet like on non fasting days.
Figure 1 Schedule intermittent fasting: F = fasting, S = study day.
The caloric intake during both diet periods was equal to avoid energy restriction
with secondary effects on metabolism. Diversion of calories between carbohydrates, fat
and protein were kept equal as well. To increase comparability, volunteers ate mainly
bread, fruits and dairy products (60% daily energy intake), supplemented with liquid
meals (40% daily energy intake): Nutridrink® (Nutricia Advanced Medical Nutrition,
Zoetermeer, the Netherlands), per serving (200ml): 300 kcal, 12.0 g protein, 36.8 g
carbohydrate and 11.6 g fat.
Furthermore, the intake of macronutrients was equal and isocaloric during both diet
periods. To succeed in finding the correct caloric need for the volunteers, resting energy
proefschrift Soeters.indb 109 4-8-2008 9:13:24
110
expenditure was measured with indirect calorimetry at inclusion. Then total energy
requirements were calculated by a dietician based on REE, dietary history and activity
score, resulting in an average of advised caloric intake of 130 - 140% of REE. We aimed
at preventing weight loss and the diets were adjusted in case of a one kilogram weight
change. However, this did not occur.
Therefore the volunteers visited the Metabolic Unit for weight control on a validated
scale (SECA 701 column scale, SECA, Hamburg, Germany) on day one of each diet and
then two times per week.
Body composition, indirect calorimetry and muscle biopsiesBody composition, oxygen consumption (VO2) and CO2 production (VCO2) were measured
as described earlier (18). Muscle biopsies were performed as described previously in the
basal state and during step 2 of the clamp (18).
Glucose, lipid and valine measurementsPlasma glucose concentrations were measured with the glucose oxidase method using
a Biosen C-line plus glucose analyzer (EKF Diagnostics, Barleben/Magdeburg, Germany).
Plasma FFA concentrations were determined with an enzymatic colorimetric method
(NEFA-C test kit, Wako Chemicals GmbH, Neuss, Germany): intra-assay variation: 1%; inter-
assay variation: 4-15%; detection limit: 0.02 mmol/L. [6,6-2H2]glucose and [1,1,2,3,3-2H5]
glycerol enrichment were determined as described earlier (18). L-[1-13C]valine, [1-13C]
α-ketoisovalerate (α-KIV) enrichments were measured as described previously (19)
Glucoregulatory hormonesInsulin was determined on an Immulite 2000 system (Diagnostic Products Corporation, Los
Angeles, USA) with a chemiluminescent immunometric assay, intra-assay variation: 3-6%,
inter-assay variation: 4-6%, detection limit: 15pmol/L. Glucagon was determined with
the Linco 125I radioimmunoassay (St. Charles, USA), intra-assay variation: 9-10%, inter-
assay variation: 5-7% and detection limit: 15ng/L. Cortisol, norepinephrine, epinephrine
and adiponectin were determined as described earlier (18).
ImmunoblottingMuscle tissue was prepared and separated by electrophoresis on a polyacrylamide gel
and immunoblots were visualized by enhanced chemi-luminescence (ECL) as described
previously (8). Chemicals for ECL were from Sigma (St.Louis, MO, USA). Phosphospecific
anti-AKTser473, phosphospecific anti-glycogen synthase kinase-3-ser9 (GSK), anti-
proefschrift Soeters.indb 110 4-8-2008 9:13:24
Intermittent fasting and glucose, lipid and protein m
etabolism
111
Chapter 7
mTORser2448 total anti-AKT and total anti-eIF4E (loading control) were from Cell
Signaling (Boston, MA, USA). Phosphospecific anti-AS160-thr642 was from GeneTex Inc.
(San Antonio, TX, USA). Results are presented as fold increase compared to control.
Calculations and statisticsEndogenous glucose production (EGP) and peripheral glucose uptake of glucose (rate
of disappearance/Rd) were calculated using the modified forms of the Steele Equations
as described previously (17;20). EGP and Rd were expressed as μmol/kg·min. Glucose
metabolic clearance rates (MCRglucose) were calculated as MCR = Ra glucose/[glucose].
Total triglyceride hydrolysis/lipolysis (glycerol turnover) was calculated by using
formulas for steady state kinetics adapted for stable isotopes (21) and was expressed as
μmol/kg·min and as μmol/kcal (18;22). Proteolysis was calculated using the reciprocal
pool model (α-KIV Ra) as described earlier (23).
REE, carbohydrate oxidation (CHO) and FAO rates were calculated from O2 consumption
and CO2 production as reported previously (24). Nonoxidative glucose disposal (NOGD)
was calculated as: NOGD = Rd - CHO.
All subjects served as their own controls. All data were analyzed with non parametric
tests. Comparisons were performed with the Wilcoxon Signed Rank test. Correlations
were expressed as Spearman’s rank correlation coefficient (ρ). The SPSS statistical
software program version 14.0.2 (SPSS Inc, Chicago, IL) was used for statistical analysis.
Data are presented as median [minimum -maximum].
Table 1 Subject characteristics at inclusion
Age (yr) 23.5 [20 - 30]
Weight (kg) 74.6 [58.4 - 85.0]
Height (cm) 188 [176 - 192]
BMI (kg/m2) 21.3 [18.2 - 24.7]
Lean body mass (%) 85.2 [75.3 - 87.9]
Fat mass (%) 14.8 [12.1 - 24.7]
FPG (mmol/liter) 5.0 [4.7 - 5.3]
OGTT (mmol/liter) 3.8 [3.4 - 5.4]
Dietary requirements
Energy requirement (kcal/day) 2653 [2395 - 2860]
Protein (gram/day) 101 [89 - 130]
Fat (gram/day) 88 [64 - 93]
Carbohydrate (gram/day) 366 [306 - 403]
Data are presented as median [minimum - maximum]. BMI, body mass index; FPG, fasting plasma glucose; OGTT, oral glucose tolerance test.
proefschrift Soeters.indb 111 4-8-2008 9:13:24
112
Results
Anthropometric characteristicsThe subject characteristics at inclusion are presented in Table 1. Weight after two weeks
of IF was not different compared to SD (74.8 [58.7 - 85.3] vs. 74.5 [58.4 - 85.7] kg
respectively, P = 0.58). Fat and lean body mass (FM and LBM) were not different after IF
compared to SD (FM 13.3 [11.7 - 20.9] % vs. 12.8 [9.9 - 15.9] % respectively, P = 0.46
and LBM [86.7 - 79.1 - 88.2] % vs. 87.3 [84.1 - 90.1] % respectively, P = 0.46).
Resting energy expenditure, glucose and fatty acid oxidationAfter two weeks of IF, REE was significantly lower in the basal state compared to two
weeks of SD: median difference of -59 [-201 - 26] kcal/day (Table 2). During the clamp,
no differences in REE were found. CHO and FAO were not different between IF and SD
during either basal state or clamp. When CHO and FAO were expressed per LBM, similar
Table 2 Indirect Calorimetry
basal state hyperinsulinaemic euglycaemic clamp
IF(n = 8)
SD(n = 8)
P IF(n = 8)
SD(n = 8)
P
REE (kcal/day) 1698 [1280 - 1754] 1716 [1478 - 1901] 0.036 1784 [1427 - 2003] 1864 [1774 - 2147] 0.33
CHO (μmol/kg•min) 4.4 [1.5 - 9.2] 6.5 [2.3 - 13.9] 0.58 19.3 [12.0 - 34.8] 20.1 [12.1 - 24.3] 0.78
FAO (μmol/kg•min) 1.5 [1.0 - 1.8] 1.4 [0.8 - 2.2] 0.89 0.4 [0 - 0.82] 0.47 [0 - 1.6] 0.48
RQ 0.76 [0.72 - 0.83] 0.79 [0.68 - 0.87] 0.67 0.93 [0.86 - 1.15] 0.93 [0.81 - 1.00] 0.78
Data are presented as median [minimum - maximum]. REE, resting energy expenditure; CHO, carbohydate oxidation; FAO, fatty acid oxidation RQ, respiratory quotient.
FIGURE 2 Differences between IF (grey box plots) and SD (open box plots) subjects in α-KIV Ra (proteolysis) during the basal state (P = 0.24) and during step 1(P = 0.13) and step 2 of the clamp, *P = 0.043. N = 7.
proefschrift Soeters.indb 112 4-8-2008 9:13:24
Intermittent fasting and glucose, lipid and protein m
etabolism
113
Chapter 7
Table 2 Indirect Calorimetry
basal state hyperinsulinaemic euglycaemic clamp
IF(n = 8)
SD(n = 8)
P IF(n = 8)
SD(n = 8)
P
REE (kcal/day) 1698 [1280 - 1754] 1716 [1478 - 1901] 0.036 1784 [1427 - 2003] 1864 [1774 - 2147] 0.33
CHO (μmol/kg•min) 4.4 [1.5 - 9.2] 6.5 [2.3 - 13.9] 0.58 19.3 [12.0 - 34.8] 20.1 [12.1 - 24.3] 0.78
FAO (μmol/kg•min) 1.5 [1.0 - 1.8] 1.4 [0.8 - 2.2] 0.89 0.4 [0 - 0.82] 0.47 [0 - 1.6] 0.48
RQ 0.76 [0.72 - 0.83] 0.79 [0.68 - 0.87] 0.67 0.93 [0.86 - 1.15] 0.93 [0.81 - 1.00] 0.78
Data are presented as median [minimum - maximum]. REE, resting energy expenditure; CHO, carbohydate oxidation; FAO, fatty acid oxidation RQ, respiratory quotient.
results were found (data not shown). Also changes during the clamp in CHO and FAO
were not different between IF and SD (data not shown).
Glucose and lipid measurementsPlasma glucose concentrations and EGP in the basal state were not statistically different
after IF compared to SD (table 3). MCRglucose in the basal state was not different between
IF and SD [2.4 (1.9 –2.8) ml/kg x min vs. 2.3 (2.0–2.9) ml/kg x min, respectively; P =
0.58]. Basal state plasma FFA levels and lipolysis after two weeks of IF were not different
from SD.
During step 1 of the clamp (Table 3), plasma glucose concentrations and EGP were
not different after IF compared to SD. Glucose Rd during step 1 was significantly higher
after IF compared to SD. Plasma FFA levels were not different during step 1, but lipolysis
tended to be lower after IF compared to SD.
Step 2 of the clamp revealed no differences in glucose Rd between IF and SD (EGP
being suppressed). Also no differences were observed in lipolysis between IF and SD
during step 2 of the clamp.
Valine measurementsKIV Ra (expressing proteolysis) was not different between IF and SD in the basal state
or step 1 of the clamp (Figure 2). KIV Ra was significantly lower after IF compared to SD
during step 2 of the clamp.
Glucoregulatory hormonesIn the basal state plasma glucagon levels were significantly lower after IF compared to
SD (Table 3). Insulin, cortisol, epinephrine, norepinephrine and adiponectin were not
different between IF and SD.
proefschrift Soeters.indb 113 4-8-2008 9:13:24
114 During step 1 of the clamp, glucagon levels tended to be lower after IF compared to
SD, whereas the other glucoregulatory hormones and adiponectin were not different.
During step 2 of the clamp, no differences were found in glucoregulatory hormones and
adiponectin except for norepinephrine that was lower after IF compared to SD (Table 3).
ImmunoblottingpAKT-ser473 in the basal state was not different between IF and SD, but during the
clamp pAKT-ser473 tended to be higher after IF compared to SD (Figure 3). pAKT-ser473
increased significantly during both clamps.
No differences were found in pAS160-thr642 in the basal state or clamp between IF
and SD (Figure 3). After both IF and SD, pAS160-thr642 tended to increase during the
clamp.
pGSK-3-ser9 was higher after IF compared to SD in the basal state as well as during the
clamp. pGSK-3-ser9 increased significantly within both clamps.
pmTOR-ser2448 tended to be lower in the basal state after IF compared to SD, whereas
pmTOR-ser2448 was significantly lower during the clamp after IF compared to SD (Figure
3). pmTOR-ser2448 increased significantly within both clamps.
Table 3 Glucose and lipid kinetics and glucoregulatory hormones
basal state clamp (step 1) clamp (step 2)
IF SD P IF SD P IF SD P
Glucose (mmol/liter) 4.8 [4.3 - 4.7] 4.7 [4.0 - 4.9] 0.09 4.9 [4.7 - 5.3] 5.0 [4.7 - 5.4] 0.83 5.2 [4.8 - 5.5] 4.9 [4.8 - 5.3] 0.14
EGP (μmol/kg•min) 11.4 [10.0 - 13.5] 10.4 [9.6 - 13.9] 0.16 2.6 [1.7 - 5.9] 1.9 [0 - 6.7] 0.67 -a -a -
Rd (μmol/kg•min) - - - 27.3 [20.4 - 36.5] 21.4 [16.5 - 35.7] 0.050 64.3 [49.4 - 89.2] 64.1 [38.9 - 81.3] 0.14
NOGD (%) - - - - - - 28.5 [20.4 - 50.5] 31.7 [19.6 - 45.2] 0.78
FFA (mmol/liter) 0.31 [0.20 - 0.44] 0.33 [0.12 - 0.55] 0.89 0.03 [0.01 - 0.07] 0.04 [0.01 - 0.14] 1 -a -a -
Lipolysis (μmol/kg•min) 1.6 [0.9 - 2.3] 1.3 [0.8 - 3.3] 0.24 0.5 [0.4 - 0.7] 0.6 [0.4 - 0.8] 0.091 0.4 [0.33 - 0.7] 0.4 [0.4 - 1.0] 0.61
Lipolysis (μmol/kcal) 104 [58 - 143] 81 [45 - 211] 0.26 - - - 27 [14 - 45] 26 [21 - 50] 0.74
Insulin (pmol/liter) 34 [<15 - 66] 32 [<15 - 63] 0.87 103 [80 - 159] 115 [87 - 149] 0.94 408 [370 - 487] 409 [344 - 436] 0.12
Glucagon (ng/liter) 46 [32 - 83] 63 [38 - 83] 0.036 39 [34 - 72] 48 [37 - 82] 0.063 33 [28 - 60] 38 [27 - 81] 0.40
Cortisol (nmol/liter) 273 [184 - 381] 291 [155 - 458] 0.67 177 [127 - 328] 201 [169 - 237] 0.33 169 [88 - 273] 76 [96 - 306] 0.48
Epinephrine (nmol/liter) 0.06 [0.03 - 0.32] 0.065 [0.03 - 0.38] 0.75 0.07 [0.03 - 0.31] 0.07 [0.03 - 0.48] 0.74 0.09 [0.03 - 0.18] 0.04 [0.03 - 0.47] 0.5
Norepinephrine (nmol/liter) 0.67 [0 - 0.98 0.60 [0.03 - 1.46] 0.67 0.60 [0.12 - 1.30] 0.79 [0.03 - 1.09] 0.4 0.48 [0.03 - 1.34] 0.69 [0.03 - 1.80] 0.025
Adiponectin (ng/ml) 8.7 [4.5 - 15.6] 7.7 [4.0 - 16.2] 0.44 8.5 [4.6 - 15.8] 7.2 [4.1 - 16.5] 0.67 7.9 [3.9 - 15.0] 6.9 [4.3 - 15.4] 0.60
Data are presented as median [minimum - maximum]. N = 8. EGP, endogenous glucose production; Rd, rate of disappearance; NOGD, non oxidative glucose disposal. aEGP and FFA were completely suppressed during step 2 of the clamp.
proefschrift Soeters.indb 114 4-8-2008 9:13:25
Intermittent fasting and glucose, lipid and protein m
etabolism
115
Chapter 7
Discussion
In this study we investigated the effect of two weeks of IF on hepatic and peripheral
insulin sensitivity (EGP and glucose uptake respectively), lipolysis and proteolysis in
comparison with two weeks of an isocaloric SD in lean healthy subjects.
Body weight and body composition did not change in the period of IF versus two
weeks of SD indicating that our findings cannot be attributed to weight loss in contrast
to a number of earlier studies (4;5). Previous reports showed no differences in data on
REE and substrate oxidation between IF and SD (1;5). Therefore, the decrease in REE after
two weeks IF that we found, is intriguing. A decreased REE may spare energy stores but
it may also increase body weight if isocaloric diets are consumed or if physical activity is
unaltered. Our study period may not have been long enough to increase body weight.
The increased peripheral insulin sensitivity during step 1 of the clamp strengthens the
data by Halberg et al. Moreover, we show that IF selectively increases peripheral insulin
sensitivity and not hepatic insulin sensitivity since we found no effect on the latter. The
reason why the insulin sensitizing effect of IF was lost during step 2 is unclear, but clamp
duration is most likely to play a role since Halberg et al reported an increase in peripheral
insulin sensitivity after a 2h clamp with an insulin infusion regimen comparable to step
2 in our study. The mechanism of the higher insulin-mediated peripheral glucose uptake
Table 3 Glucose and lipid kinetics and glucoregulatory hormones
basal state clamp (step 1) clamp (step 2)
IF SD P IF SD P IF SD P
Glucose (mmol/liter) 4.8 [4.3 - 4.7] 4.7 [4.0 - 4.9] 0.09 4.9 [4.7 - 5.3] 5.0 [4.7 - 5.4] 0.83 5.2 [4.8 - 5.5] 4.9 [4.8 - 5.3] 0.14
EGP (μmol/kg•min) 11.4 [10.0 - 13.5] 10.4 [9.6 - 13.9] 0.16 2.6 [1.7 - 5.9] 1.9 [0 - 6.7] 0.67 -a -a -
Rd (μmol/kg•min) - - - 27.3 [20.4 - 36.5] 21.4 [16.5 - 35.7] 0.050 64.3 [49.4 - 89.2] 64.1 [38.9 - 81.3] 0.14
NOGD (%) - - - - - - 28.5 [20.4 - 50.5] 31.7 [19.6 - 45.2] 0.78
FFA (mmol/liter) 0.31 [0.20 - 0.44] 0.33 [0.12 - 0.55] 0.89 0.03 [0.01 - 0.07] 0.04 [0.01 - 0.14] 1 -a -a -
Lipolysis (μmol/kg•min) 1.6 [0.9 - 2.3] 1.3 [0.8 - 3.3] 0.24 0.5 [0.4 - 0.7] 0.6 [0.4 - 0.8] 0.091 0.4 [0.33 - 0.7] 0.4 [0.4 - 1.0] 0.61
Lipolysis (μmol/kcal) 104 [58 - 143] 81 [45 - 211] 0.26 - - - 27 [14 - 45] 26 [21 - 50] 0.74
Insulin (pmol/liter) 34 [<15 - 66] 32 [<15 - 63] 0.87 103 [80 - 159] 115 [87 - 149] 0.94 408 [370 - 487] 409 [344 - 436] 0.12
Glucagon (ng/liter) 46 [32 - 83] 63 [38 - 83] 0.036 39 [34 - 72] 48 [37 - 82] 0.063 33 [28 - 60] 38 [27 - 81] 0.40
Cortisol (nmol/liter) 273 [184 - 381] 291 [155 - 458] 0.67 177 [127 - 328] 201 [169 - 237] 0.33 169 [88 - 273] 76 [96 - 306] 0.48
Epinephrine (nmol/liter) 0.06 [0.03 - 0.32] 0.065 [0.03 - 0.38] 0.75 0.07 [0.03 - 0.31] 0.07 [0.03 - 0.48] 0.74 0.09 [0.03 - 0.18] 0.04 [0.03 - 0.47] 0.5
Norepinephrine (nmol/liter) 0.67 [0 - 0.98 0.60 [0.03 - 1.46] 0.67 0.60 [0.12 - 1.30] 0.79 [0.03 - 1.09] 0.4 0.48 [0.03 - 1.34] 0.69 [0.03 - 1.80] 0.025
Adiponectin (ng/ml) 8.7 [4.5 - 15.6] 7.7 [4.0 - 16.2] 0.44 8.5 [4.6 - 15.8] 7.2 [4.1 - 16.5] 0.67 7.9 [3.9 - 15.0] 6.9 [4.3 - 15.4] 0.60
Data are presented as median [minimum - maximum]. N = 8. EGP, endogenous glucose production; Rd, rate of disappearance; NOGD, non oxidative glucose disposal. aEGP and FFA were completely suppressed during step 2 of the clamp.
proefschrift Soeters.indb 115 4-8-2008 9:13:25
116
FIGURE 3 Panel A pAKT-ser473 (n = 7) in the basal state (P = 1.0) and during the hyperinsulinemic euglycemic clamp (**P = 0.063) after intermittent fasting (IF) (open box plots) and standard diet (SD) (grey box blots). *P = 0.018 and 0.028 for the difference in pAKT-ser473 between basal and clamp after IF and SD respectively.Panel B pAS160-thr642 (n = 7) in the basal state (P = 0.55) and during the hyperinsulinemic euglycemic clamp (P = 0.45) after IF (open box plots) and SD (grey box blots). **P = 0.091 and 0.063 for the difference in pAS160-thr642 between basal and clamp after IF and SD respectively.Panel C pGSK-3-ser9 (n = 7) in the basal state (**P = 0.063) and during the hyperinsulinemic euglycemic clamp (P = 0.018) after IF (open box plots) and SD (grey box blots). *P = 0.018 and 0.018 for the difference in pGSK-3-ser9 between basal and clamp after IF and SD respectively.Panel D pmTOR-ser2448 (n = 7) in the basal state tended to be lower after IF (open box plots) compared to SD (grey box blots) (**P = 0.063) whereas pmTOR-ser2448 was significantly lower during the hyperinsulinemic euglycemic clamp (P = 0.028) after IF compared to SD.
proefschrift Soeters.indb 116 4-8-2008 9:13:25
Intermittent fasting and glucose, lipid and protein m
etabolism
117
Chapter 7
remains to be elucidated. Glucoregulatory hormones were not different except for
glucagon. However, glucagon seems to have its major effect on EGP and not Rd (25).
FFA may be an alternative factor in modulating peripheral insulin sensitivity since
the interfering role of FFA and their metabolites with the insulin signaling cascade is
generally appreciated (26;27). It was shown in mice that IF induces an increase in plasma
FFA levels on the one hand and smaller adipocytes on the other suggesting increased
triglyceride hydrolysis, but these animals lost weight during the study period (28). The
authors questioned whether such increase in plasma FFA levels would be a beneficial
adaptation. Although the 20 h fasting periods in our study may have been long enough
to stimulate lipolysis and FAO (29), we did not detect changes in lipolysis or plasma FFA in
the basal state after an overnight fast, thereby confirming previous human studies (1;5).
During step 1 of the clamp, lipolysis (expressed as μmol/kg x min) tended to be lower
after IF compared with SD in keeping with the increased suppressibility of adipose tissue
lipolysis after IF as shown earlier by Halberg et al (1). Although it is attractive to explain
the increased insulin sensitivity after IF by lower lipolysis and thereby lower exposure to
FFA, we cannot make a definite assumption.
Earlier we showed that decreased peripheral insulin sensitivity during short-term
fasting in lean healthy humans is explained by lower phosphorylation of muscle AKT-
ser473 and AS160-thr642 under hyperinsulinemic circumstances (8). We only measured
phosphorylation of these signaling proteins during step 2 of the clamp and not during
step 1 when peripheral insulin sensitivity differed between IF and SD. The trend towards
higher pAKT-ser473 suggests that changes in muscle insulin signaling are present after IF
in lean healthy individuals that favor peripheral glucose uptake although we cannot make
a definitive assumption. Moreover, this trend was not accompanied by a difference in
pAS160-thr642, a downstream target of AKT that is involved in the translocation of GLUT4
to the plasma membrane. Insulin mediated phosphorylation of AS160 was shown to be
decreased in patients with type 2 diabetes (18). Another downstream target of AKT is
GSK that, once phosphorylated by AKT, stimulates glycogen synthesis (20). We showed
previously that short-term fasting does not influence phosphorylation of GSK-3-ser9 in the
basal state nor during hyperinsulinemia (1). However, the higher basal and clamp pGSK-
3-ser9 levels after IF suggest adaptation that favors glycogen replenishment.
The adipokine adiponectin is thought to increase insulin sensitivity and fatty acid
oxidation and is decreased in insulin resistant states as obesity and type 2 diabetes mellitus
(30). Also, adiponectin was suggested to play a role in increased insulin sensitivity in IF
although adiponectin levels were only increased at the end of fasting days and not in the
basal state (1). In contrast it was shown that plasma adiponectin levels during short-term
proefschrift Soeters.indb 117 4-8-2008 9:13:25
118
fasting do not change within days when more frequent sampling was performed (31).
The lack of change in adiponectin levels and FAO between IF and SD does not support a
role for adiponectin in increasing peripheral insulin sensitivity during IF.
Protein metabolism in IF has not been investigated in humans before to our knowledge.
Intracellular valine contributes to the citric acid cycle intermediate succinyl-CoA (via
transamination to α-KIV that is converted to propionyl-CoA). α-KIV Ra is assumed to
reflect the intracellular valine enrichment and proteolysis (32). We show that IF lowers
α-KIV Ra during step 2 of the clamp. Collectively these data suggest that IF does not
decrease proteolysis in the basal state but increases insulin mediated inhibition of
proteolysis during higher insulin levels.
mTOR is involved in protein synthesis and activated by growth factors and nutrient
sensitive pathways (e.g. amino acids) (33). Although we did not assess protein synthetic
rates, the decrease in mTOR phosphorylation suggests that IF lowers protein synthesis as
shown earlier for short-term fasting (14). Whether our results indicate maintenance of
skeletal muscle mass during IF remains elusive.
In conclusion, IF increases insulin-mediated peripheral glucose uptake but not hepatic
insulin sensitivity. Insulin sensitivity of adipose tissue tends to be augmented whereas
insulin-mediated suppression of proteolysis was significantly higher. The observed
changes in glucose fluxes may be mediated via changes in phosphorylation of muscle
AKT and GSK. The decrease in REE after IF may precede weight gain during IF when
caloric intake is not adjusted. Whether IF is beneficial in improving peripheral insulin
resistance in obese, insulin resistant subjects remains to be established.
AcknowledgementsThe authors would like to thank A.F.C Ruiter for excellent assistance on stable isotope
analyses, M. van Vessem for contributing to the experimental work and Nutricia Advanced
Medical Nutrition (Zoetermeer, the Netherlands) for kindly providing Nutridrink®.
Reference List 1. Halberg N, Henriksen M, Soderhamn N et al. Effect of intermittent fasting and refeeding on
insulin action in healthy men. J Appl Physiol 2005; 99(6):2128-2136.
2. Chakravarthy MV, Booth FW. Eating, exercise, and “thrifty” genotypes: connecting the dots toward an evolutionary understanding of modern chronic diseases. J Appl Physiol 2004; 96(1):3-10.
3. Neel JV. Diabetes mellitus: a “thrifty” genotype rendered detrimental by “progress”? Am J Hum Genet 1962; 14:353-62.:353-362.
proefschrift Soeters.indb 118 4-8-2008 9:13:25
Intermittent fasting and glucose, lipid and protein m
etabolism
119
Chapter 7
4. Heilbronn LK, Civitarese AE, Bogacka I, Smith SR, Hulver M, Ravussin E. Glucose tolerance and skeletal muscle gene expression in response to alternate day fasting. Obes Res 2005; 13(3):574-581.
5. Heilbronn LK, Smith SR, Martin CK, Anton SD, Ravussin E. Alternate-day fasting in nonobese subjects: effects on body weight, body composition, and energy metabolism. Am J Clin Nutr 2005; 81(1):69-73.
6. Klein S, Sakurai Y, Romijn JA, Carroll RM. Progressive alterations in lipid and glucose metabolism during short-term fasting in young adult men. Am J Physiol Endocrinol Metab 1993; 265(5):E801-E806.
7. Bergman BC, Cornier MA, Horton TJ, Bessesen DH. Effects of fasting on insulin action and glucose kinetics in lean and obese men and women. Am J Physiol Endocrinol Metab 2007; 293(4):E1103-E1111.
8. Soeters MR, Sauerwein HP, Dubbelhuis PF et al. Muscle adaptation to short-term fasting in healthy lean humans. J Clin Endocrinol Metab 2008;jc.
9. Bjorkman O, Eriksson LS. Influence of a 60-hour fast on insulin-mediated splanchnic and peripheral glucose metabolism in humans. J Clin Invest 1985; 76(1):87-92.
10. Fery F, d’Attellis NP, Balasse EO. Mechanisms of starvation diabetes: a study with double tracer and indirect calorimetry. Am J Physiol Endocrinol Metab 1990; 259(6):E770-E777.
11. van der Crabben SN, Allick G, Ackermans MT, Endert E, Romijn JA, Sauerwein HP. Prolonged Fasting Induces Peripheral Insulin Resistance, Which Is Not Ameliorated by High-Dose Salicylate. J Clin Endocrinol Metab 2008; 93(2):638-641.
12. Fryburg DA, Barrett EJ, Louard RJ, Gelfand RA. Effect of starvation on human muscle protein metabolism and its response to insulin. Am J Physiol Endocrinol Metab 1990; 259(4):E477-E482.
13. Tsalikian E, Howard C, Gerich JE, Haymond MW. Increased leucine flux in short-term fasted human subjects: evidence for increased proteolysis. Am J Physiol Endocrinol Metab 1984; 247(3):E323-E327.
14. Afolabi PR, Jahoor F, Jackson AA et al. The effect of total starvation and very low energy diet in lean men on kinetics of whole body protein and five hepatic secretory proteins. Am J Physiol Endocrinol Metab 2007; 293(6):E1580-E1589.
15. Lecker SH, Goldberg AL, Mitch WE. Protein Degradation by the Ubiquitin-Proteasome Pathway in Normal and Disease States. J Am Soc Nephrol 2006; 17(7):1807-1819.
16. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 2003; 26(90001):5S-20.
17. Finegood DT, Bergman RN, Vranic M. Estimation of endogenous glucose production during hyperinsulinemic-euglycemic glucose clamps. Comparison of unlabeled and labeled exogenous glucose infusates. Diabetes 1987; 36(8):914-924.
18. Soeters MR, Sauerwein HP, Groener JE et al. Gender-Related Differences in the Metabolic Response to Fasting. J Clin Endocrinol Metab 2007; 92(9):3646-3652.
19. Geukers VG, Oudshoorn JH, Taminiau JA et al. Short-term protein intake and stimulation of protein synthesis in stunted children with cystic fibrosis. Am J Clin Nutr 2005; 81(3):605-610.
20. STEELE R. Influences of glucose loading and of injected insulin on hepatic glucose output. Ann N Y Acad Sci 1959; 82:420-30.:420-430.
21. Ackermans MT, Ruiter AF, Endert E. Determination of glycerol concentrations and glycerol isotopic enrichments in human plasma by gas chromatography/mass spectrometry. Anal Biochem 1998; 258(1):80-86.
22. Koutsari C, Jensen MD. Thematic review series: patient-oriented research. Free fatty acid metabolism in human obesity. J Lipid Res 2006; 47(8):1643-1650.
23. Bisschop PH, de Sain-van der Velden, Stellaard F et al. Dietary Carbohydrate Deprivation Increases 24-Hour Nitrogen Excretion without Affecting Postabsorptive Hepatic or Whole Body Protein Metabolism in Healthy Men. J Clin Endocrinol Metab 2003; 88(8):3801-3805.
proefschrift Soeters.indb 119 4-8-2008 9:13:25
120
24. Frayn KN. Calculation of substrate oxidation rates in vivo from gaseous exchange. J Appl Physiol 1983; 55(2):628-634.
25. Barthel A, Schmoll D. Novel concepts in insulin regulation of hepatic gluconeogenesis. Am J Physiol Endocrinol Metab 2003; 285(4):E685-E692.
26. Boden G, Lebed B, Schatz M, Homko C, Lemieux S. Effects of Acute Changes of Plasma Free Fatty Acids on Intramyocellular Fat Content and Insulin Resistance in Healthy Subjects. Diabetes 2001; 50(7):1612-1617.
27. Holland WL, Knotts TA, Chavez JA, Wang LP, Hoehn KL, Summers SA. Lipid Mediators of Insulin Resistance. Nutrition Reviews 2007; 65:39-46.
28. Varady KA, Roohk DJ, Loe YC, Evoy-Hein BK, Hellerstein MK. Effects of modified alternate-day fasting regimens on adipocyte size, triglyceride metabolism, and plasma adiponectin levels in mice. J Lipid Res 2007; 48(10):2212-2219.
29. Corssmit EPM, Stouthard JML, Romijn JA, Endert E, Sauerwein HP. Sex differences in the adaptation of glucose metabolism to short-term fasting: Effects of oral contraceptives. Metabolism 1994; 43(12):1503-1508.
30. Ahima RS. Adipose Tissue as an Endocrine Organ. Obesity 2006; 14(S8):242S-249S.
31. Merl V, Peters A, Oltmanns KM et al. Serum adiponectin concentrations during a 72-hour fast in over- and normal-weight humans. Int J Obes (Lond) 2005; 29(8):998-1001.
32. Horber FF, Horber-Feyder CM, Krayer S, Schwenk WF, Haymond MW. Plasma reciprocal pool specific activity predicts that of intracellular free leucine for protein synthesis. Am J Physiol Endocrinol Metab 1989; 257(3):E385-E399.
33. Dann SG, Thomas G. The amino acid sensitive TOR pathway from yeast to mammals. FEBS Letters 2006; 580(12):2821-2829.
proefschrift Soeters.indb 120 4-8-2008 9:13:25
proefschrift Soeters.indb 121 4-8-2008 9:13:25
proefschrift Soeters.indb 122 4-8-2008 9:13:25
Hypoinsulinemic hypoglycemia during fasting in adults: a Gaussian Tale8Maarten R. Soeters1, Hidde H. Huidekoper2,
Marinus Duran3, Mariëtte T. Ackermans4, Erik
Endert4, Eric Fliers1, Frits A, Wijburg2, Hans P.
Sauerwein1 and Mireille J. Serlie1
1Department of Endocrinology and Metabolism2Department of Pediatrics3Department of Clinical Chemistry, Laboratory
Genetic Metabolic Diseases4Department of Clinical Chemistry, Laboratory
of Endocrinology Academic Medical Center,
University of Amsterdam, Amsterdam, the
Netherlands
Submitted
proefschrift Soeters.indb 123 4-8-2008 9:13:25
Abstract
Background: The diagnostic evaluation of spontaneous hypoglycemia is mainly directed at
detecting an insulinoma, via a prolonged supervised fast. Its interpretation is troublesome
in those patients who develop hypoglycemia with appropriate hypoinsulinemia during a
prolonged supervised fast. In this study we investigated in this group of patients whether
abnormalities in intermediary metabolism (fatty acid oxidation and amino/organic acids)
could be detected which might explain the hypoinsulinemic hypoglycemia.
Methods: 10 patients with otherwise unexplained hypoglycemia during prolonged
fasting participated in an extended metabolic diagnostic protocol based on stable isotope
techniques after an overnight fast in order to explore abnormalities in endogenous
glucose production (EGP) and intermediary metabolism.
Results: Although during the prolonged fast, they became hypoglycaemic, during the
diagnostic protocol there were no hypoglycemic events. No abnormalities in fatty acid
oxidation (FAO) or in amino acid/organic acids were found in this patient group.
Conclusion: In patients exhibiting hypoglycemia during prolonged fasting in whom
insulinoma was excluded, we found no signs of metabolic derangements. Therefore,
the previously observed low plasma glucose values in this subgroup of patients probably
represent the lower tail of the Gaussian Curve of plasma glucose concentrations during
fasting.
proefschrift Soeters.indb 124 4-8-2008 9:13:25
Hypoinsulinem
ic hypoglycemia: a G
aussian Tale
125
Chapter 8
Introduction
The diagnostic evaluation of spontaneous hypoglycemia can be troublesome (1). A
prolonged supervised fast is a validated clinical test to confirm and explore suspected
hypoglycemia (2). However, a substantial part of the evaluated patients has plasma
glucose levels that fall well below 3.0 mmol/liter in the absence of hyperinsulinemia and
signs of neurohypoglycemia (unpublished observation).
Plasma glucose levels decrease during fasting, notwithstanding the integrated
metabolic response that prevents energy depletion and protects the central nervous
system from hypoglycemia (3-6). This response consists amongst others of increased fatty
acid oxidation (FAO) to spare glucose consumption. During fasting, the combination of
low plasma insulin levels and increased levels of cortisol, growth hormone (GH), glucagon
and catecholamines increase lipolysis of adipose tissue thereby increasing the plasma free
fatty acid (FFA) concentration and subsequent FAO in oxidative tissues (e.g. skeletal
muscle) (7;8). Additionally, the increased FFA availability in the liver stimulates ketone
body (KB) formation (6;9).
The decrease in plasma glucose levels during fasting is, in adults, mainly due to a
decrease in endogenous glucose production (EGP) (5) which is the main denominator of
the fasting plasma glucose level (10). Despite the inevitable lowering of plasma glucose
concentration during fasting, levels normally do not fall below 3.0 mmol/liter (5;6).
However, healthy women may have plasma glucose levels well below 3.0 mmol/liter
during fasting (5;6;11). In addition, in a number of inborn errors of metabolism, including
glycogen storage diseases, disorders of gluconeogenesis and disorders of mitochondrial
FAO marked fasting hypoglycemia is one of the most important biochemical signs (12).
Medium-chain acyl-CoA dehydrogenase deficiency (MCADD, OMIM# 201450) is
probably the most common FAO disorder in which fasting induced FAO results in the
accumulation of toxic FAO intermediates and their CoA esters (12;13). The latter are
thought to contribute to the clinical presentation of FAO disorders: MCADD patients
present with hypoketotic hypoglycemia (decreased EGP) despite hypo-insulinemia during
episodes of fasting (12;13). A number of other inborn defects of other, chain length
specific, acyl-CoA dehydrogenases, catalyzing the first step in intramitochondrial FAO,
have now been characterized. All of these may present with hypoketotic hypoglycemia
(14). Since the accumulating acyl-CoA esters are converted to their matching acylcarnitine
esters and released in plasma; analysis of these plasma acylcarnitine profiles by tandem
proefschrift Soeters.indb 125 4-8-2008 9:13:25
126
mass spectrometry is the current standard for the diagnosis of FAO disorders at the
metabolite level (14).
In this descriptive report, we present 10 patients with hypoinsulinemic hypoglycemia,
in whom insulinoma had been excluded. The patients now participated in an extended
metabolic diagnostic protocol designed to explore whether abnormalities in EGP
(measured with the stable isotope technique), plasma amino/organic acids (defective
gluconeogenesis and glycogenolysis) or FAO (whole body FAO and plasma acylcarnitine
profiles) could explain the clinical findings.
Subjects and methods
SubjectsBetween 2000 and 2007 48 patients underwent a prolonged supervised fast because of
hypoglycemic complaints. Of these patients, 33% (n = 16) had plasma glucose levels during
the prolonged supervised fast below 3.0 mmol/liter with concomitant hypoinsulinemia
(insulin < 42 pmol/L) and no Whipple’s triad. Of these patients, we included 10 patients.
The other patients declined to participate or their physician did not refer them to the
Metabolic Unit.
The limit of 3.0 mmol/liter was set because β-cell polypeptide secretion is suppressed
below that glucose level (2). Patients had undergone the prolonged test in our center
or were referred by other hospitals because of unexplained hypoglycemia. Other criteria
for inclusion in our study were: 1) no medication; 2) absence of diabetes or insulin
administration; 3) absence of adrenocortical failure, which was excluded by an ACTH-
stimulation test (15). Patients were informed on the purposes and nature of the diagnostic
protocol.
Experimental protocolFor three days before the diagnostic protocol, patients consumed a weight-maintaining
diet containing at least 250 g of carbohydrates per day. Patients were studied after 16
hours of fasting: patients were fasting from 1800 h the evening before the study day
until the end of the study day. They were allowed to drink water only. After admission
at the Metabolic Unit of the Academic Medical Center of the University of Amsterdam
at 0730 h, a catheter was inserted into an antecubital vein for infusion of the stable
isotope tracer. Another catheter was inserted retrogradely into a contralateral hand vein
and kept in a thermo-regulated (60ºC) plexiglas box for sampling of arterialized venous
proefschrift Soeters.indb 126 4-8-2008 9:13:25
Hypoinsulinem
ic hypoglycemia: a G
aussian Tale
127
Chapter 8
blood. In all studies, saline was infused as NaCl 0.9% at a rate of 50 mL/h to keep
the catheters patent, [6.6-2H2]glucose was used as tracer (>99% enriched; Cambridge
Isotopes, Andover, USA) to study glucose kinetics.
At 0800 h, blood samples were drawn for determination of background enrichments
and a primed continuous infusion of [6.6-2H2]glucose was started: prime, 8.8 μmol/
kg; continuous, 0.11 μmol/kg·min and continued until the end of the study. After an
equilibration period of two hours (16 h of fasting), 3 blood samples were drawn for glucose
enrichments and 1 for glucoregulatory hormones, FFA, KBs, alanine and acylcarnitines.
Then blood glucose levels were measured every 30 minutes at the bedside to guard
possible developing hypoglycemia. After 6 h (22 h of fasting) again 3 blood samples were
drawn for glucose enrichments and 1 for glucoregulatory hormones, FFA, KBs, alanine
and acylcarnitines after which the test ended. The diagnostic protocol was discontinued
when plasma glucose levels fell below 2.5 mmol/liter and/or neurohypoglycemia
developed. In such case, an intravenous glucose solution (25 grams) was administered
and the patient was fed oral carbohydrates.
Indirect calorimetry (glucose and fat oxidation)Glucose and fat oxidation were measured after 16 and 22 h of fasting during 20 min
by indirect calorimetry using a ventilated hood system (Sensormedics model 2900;
Sensormedics, Anaheim, USA).
Plasma glucose, FFA, ketone bodies, lactatePlasma glucose concentrations were measured with the glucose oxidase method using
a Beckman Glucose Analyzer 2 (Beckman, Palo Alto, USA, intra-assay variation 2-3%).
[6.6-2H2]glucose enrichment was measured as described earlier (16). [6.6-2H2]glucose
enrichment (tracer/tracee ratio) intra-assay variation: 0.5-1%; inter-assay variation 1%;
detection limit: 0.04%.
Plasma FFA concentrations were determined with an enzymatic colorimetric method
(NEFA-C test kit, Wako Chemicals GmbH, Neuss, Germany): intra-assay variation 1%;
inter-assay variation: 4-15%; detection limit: 0.02 mmol/liter. KB samples were drawn in
chilled sodium-fluoride tubes and directly deproteinized with ice cold perchloric acid 6%
(1:1). AcAc and D-3HB were measured with an enzymatic/spectrophotometric method
using D-3-hydroxybutyrate dehydrogenase (COBAS-FARA centrifugal analyzer, Roche
Diagnostics, Almere, the Netherlands), Detection limit AcAc 0.05 mmol/liter; D-3HB 0.1
mmol/liter.
proefschrift Soeters.indb 127 4-8-2008 9:13:25
128
Glucoregulatory hormonesInsulin and cortisol were determined on an Immulite 2000 system (Diagnostic Products
Corporation, Los Angeles, USA). Insulin was measured with a chemiluminescent
immunometric assay, intra-assay variation: 3-6%, inter-assay variation: 4-6%, detection
limit: 15pmol/L. Cortisol was measured with a chemiluminescent immunoassay, intra-
assay variation: 7-8%, inter-assay variation: 7-8%, detection limit: 50nmol/L. Glucagon
was determined with the Linco 125I radioimmunoassay (St. Charles, USA), intra-assay
variation: 9-10%, inter-assay variation: 5-7% and detection limit: 15ng/L. Norepinephrine
and epinephrine were determined with an in-house HPLC method, Intra-assay variation
norepinephrine: 2%; epinephrine 9%; inter-assay variation norepinephrine: 10%;
epinephrine: 14 -18%; detection limit: 0.05 nmol/L. Presented reference values are the
reference values of the Laboratory of Endocrinology of the Academical Medical Center.
Plasma acylcarnitinesFree carnitine and short- (C2 and C4), medium- (C8) and long-chain (C14:1, C16:1 and
C18:1) acylcarnitines were measured by a quantitative analysis using electrospray tandem
mass spectrometry as described earlier (17); intra-assay variation 6-15% for free carnitine
as well as the different acylcarnitines.
Organic and amino acid analysisOrganic acids (e.g. methylmalon acid and glutaric acid) in plasma were analyzed by
conventional gas chromatography/mass spectrometry following extraction with ethyl
acetate/diethyl ether. The acids were separated as their methoxime/trimethylsilyl esters.
Amino acids in plasma were analyzed (e.g. alanine and glutamine) using reversed-phase
HPLC combined with electrospray tandem mass spectrometry of the underivatized amino
acid. Reference values of the separate organic and amino acids were in-house reference
values of the Laboratory Genetic Metabolic Diseases of the AMC.
Calculations and statisticsEGP was calculated as the rate of isotope infusion (μmol/kg·min) divided by plateau tracer
tracee ratios (TTR), hence EGP was expressed as μmol/kg·min. To compare the EGP of
patients to a reference population, data on EGP were used from previously studies in
healthy lean volunteers (n = 63, partly unpublished) after 14-16 h of fasting (4;16;18-23).
A central range, encompassing 95% of the data values was set, using the 2.5th and
97.5th percentiles as limits.
proefschrift Soeters.indb 128 4-8-2008 9:13:25
Hypoinsulinem
ic hypoglycemia: a G
aussian Tale
129
Chapter 8
Glucose metabolic clearance rates (MCRglucose) were calculated as EGP/[glucose].
Glucose and fat oxidation rates were calculated from O2 consumption and CO2 production
as reported previously (24).
Data are presented as individual results and as median [minimum - maximum] when
expressed for the total group. Comparisons were performed with the Wilcoxon Signed
Rank test. The SPSS statistical software program version 14.0.2 (SPSS Inc, Chicago, IL)
was used for statistical analysis.
Results
Anthropometric characteristicsTen patients agreed to participate in the diagnostic protocol. The subject characteristics
are presented in Table 1. Women outnumbered men (9 vs. 1 respectively). The median
age of the patients was 40 [21 – 57] yr. As a group, patients had a normal BMI: 24 [17.7
– 29.4] kg/m2.
Table 1 Patient characteristics prior to inclusion
Sex Age(yr)
Height(cm)
Weight(kg)
BMI(kg/m2)
Glucose(mmol/liter)
Insulin(pmol/liter)
Patient 1 v 43 158 67 26.8 2.6 <15a
Patient 2 v 29 170 85 29.4 2.8 20
Patient 3 v 21 174 69 22.8 2.5 16
Patient 4 v 46 147 51 23.4 2.5 <15a
Patient 5 v 40 178 56 17.7 2.4 22
Patient 6 v 35 157 59 23.9 2.5 20
Patient 7 v 57 163 63 23.7 2.1 <15a
Patient 8 m 38 172 70 23.7 2.9 <15a
Patient 9 v 43 182 70 21.1 2.9 <15a
Patient 10 v 40 178 73 23.0 2.1 <15a
Data are presented as individual values. aPlasma insulin levels were below the detection limit.
Glucose kineticsAll patients had low plasma glucose levels during the prolonged supervised fast with low
plasma insulin levels (Table 1). Plasma glucose levels decreased in all patients between 16
and 22 hours of fasting with a median decrease of 12 [2.9 - 25.8] %, P = 0.005 (Table 2).
All patients had an EGP within our reference values (reference values for EGP after 14-16
h of fasting: 7.98 - 14.46 μmol/kg·min) (4;18-20;22;25). EGP decreased in all patients
proefschrift Soeters.indb 129 4-8-2008 9:13:26
130between 16 and 22 hours of fasting with a median decrease of 18.3 [13.0 - 29.1] %,
P = 0.005. MCRglucose was lower after 22 h of fasting compared to 16 h of fasting in 9
patients, but did not change in patient 8 (median decrease 14 [0 - 18] %, P = 0.007 for
the whole group).
Table 2 Glucose kinetics
Glucose EGPa MCRglucose (ml/kg · min)
(mmol/liter) Δ (%) (µmol/kg · min) Δ (%)
16h 22h 16h 22h 16h 22h
Patient 1 4.9 4.1 -16.5 9.2 6.6 -29.1 1.9 1.6
Patient 2 5.2 4.7 -8.3 8.4 7.0 -16.5 1.6 1.5
Patient 3 5.1 4.5 -12.1 10.9 8.9 -18.1 2.1 2,0
Patient 4 4.9 4.1 -15.3 12.1 9.9 -18.5 2.5 2.4
Patient 5 4.5 4.4 -2.9 11.2 9.3 -16.6 2.5 2.1
Patient 6 4.8 4.5 -5.6 10.2 8.1 -20.0 2.1 1.8
Patient 7 4.5 4.4 -2.9 9.8 7.9 -19.8 2.2 1.8
Patient 8 5.3 4.6 -12.5 11.4 9.9 -13.0 2.2 2.2
Patient 9 4.9 4.7 -4.1 11.6 9.6 -17.4 2.4 2,0
Patient 10 4.2 3.7 -11.6 11.2 9.1 -18.9 2.7 2.5
Data are presented as individual values. EGP endogenous glucose production. aReference values EGP after 16 h of fasting: 7.98 - 14.46 μmol/kg•min.
Table 3 Plasma FFA, ketone bodies, lactate and alanine
FFA (mmol/liter) AcAc (mmol/liter) D-3BH (mmol/liter) Lactate (mmol/liter) Alanine (µmol/liter)
Reference values 182 - 552
16h 22h 16h 22h 16h 22h 16h 22h 16h 22h
Patient 1 0.56 0.8 0.09 0.13 0.1 0.4 0.6 0.6 199 162
Patient 2 0.66 0.84 <0.05a 0.07 0.1 0.1 0.6 0.4 280 232
Patient 3 1.12 0.82 0.09 0.12 0.2 0.3 0.8 0.8 236 287
Patient 4 0.7 0.99 0.06 0.24 0.1 0.5 0.5 0.6 159 137
Patient 5 0.84 0.7 0.05 0.25 0.1 0.6 1.1 0.7 161 121
Patient 6 0.49 0.79 0.05 0.21 0.1 0.6 <0.5a 0.6 213 202
Patient 7 0.66 0.97 0.18 0.31 0.3 0.7 0.6 0.6 148 138
Patient 8 0.33 0.44 <0.05a <0.05a <0.1a <0.1a 0.7 0.6 211 191
Patient 9 0.66 0.57 0.08 0.13 0.3 0.4 0.5 0.6 134 135
Patient 10 0.93 0.82 <0.05a 0.18 <0.1a 0.5 0.7 0.6 165 133
Data are presented as individual values. FFA free fatty acids, AcAc acetoacetate, D-3HB D-3hydroxybutyrate. aBelow detection limit.
proefschrift Soeters.indb 130 4-8-2008 9:13:26
Hypoinsulinem
ic hypoglycemia: a G
aussian Tale
131
Chapter 8
Plasma FFA, ketone bodies, and lactatePlasma FFA levels were detectable in all patients after 16 h of fasting (Table 3). For the
group as a whole, no change in plasma FFA levels was observed (P = 0.2). During the
diagnostic protocol AcAc increased (median increase 72 [0 - 400] %) in all patients but
one; D-3HB increased (median increase 133 [0 - 900] %) in all but two patients (P = 0.008
and 0.011 for AcAc and D-3HB respectively). One of the patients did not display ketosis
at all (patient 8). All plasma lactate levels were within hospital reference values after 16
as well as 22 h of fasting and did not change (P = 0.7).
Glucoregulatory hormonesPlasma insulin levels were detectable in all patients after 16 h of fasting and decreased
thereafter (median decrease 47 [10 - 78] %), P = 0.005 (Table 4). Plasma glucagon levels
increased (median increase 18 [-12 - 28] %) between 16 and 22 h of fasting in all but
one patient, P = 0.038. Plasma cortisol levels were within the reference values normal
reference range for fasting cortisol (15) and decreased (median decrease 20 [-68 - 58] %)
during the day, P = 0.037. Plasma noradrenaline levels were within the reference range
in all patients and decreased (median decrease 27 [-8 - 46] %) for the group as a whole,
P = 0.017. Plasma adrenalin levels were within the reference range in all patients without
significant change between 16 and 22 h of fasting (P = 0.6).
Table 3 Plasma FFA, ketone bodies, lactate and alanine
FFA (mmol/liter) AcAc (mmol/liter) D-3BH (mmol/liter) Lactate (mmol/liter) Alanine (µmol/liter)
Reference values 182 - 552
16h 22h 16h 22h 16h 22h 16h 22h 16h 22h
Patient 1 0.56 0.8 0.09 0.13 0.1 0.4 0.6 0.6 199 162
Patient 2 0.66 0.84 <0.05a 0.07 0.1 0.1 0.6 0.4 280 232
Patient 3 1.12 0.82 0.09 0.12 0.2 0.3 0.8 0.8 236 287
Patient 4 0.7 0.99 0.06 0.24 0.1 0.5 0.5 0.6 159 137
Patient 5 0.84 0.7 0.05 0.25 0.1 0.6 1.1 0.7 161 121
Patient 6 0.49 0.79 0.05 0.21 0.1 0.6 <0.5a 0.6 213 202
Patient 7 0.66 0.97 0.18 0.31 0.3 0.7 0.6 0.6 148 138
Patient 8 0.33 0.44 <0.05a <0.05a <0.1a <0.1a 0.7 0.6 211 191
Patient 9 0.66 0.57 0.08 0.13 0.3 0.4 0.5 0.6 134 135
Patient 10 0.93 0.82 <0.05a 0.18 <0.1a 0.5 0.7 0.6 165 133
Data are presented as individual values. FFA free fatty acids, AcAc acetoacetate, D-3HB D-3hydroxybutyrate. aBelow detection limit.
proefschrift Soeters.indb 131 4-8-2008 9:13:26
132
Glucose and fat oxidationGlucose oxidation rates were lower after 22 h compared to 16 h of fasting (median
decrease 55 [-66 - 87] %) in all subjects but one (patient 7, Table 5), P = 0.011. Likewise,
fat oxidation rates were increased (median increase 72 [-12 - 150] %) after 22 h compared
to 16 h of fasting in all subjects but one (patient 7), P = 0.015.
Table 4 Glucoregulatory hormones
Insulin (pmol/liter)
Glucagon (ng/liter)
Cortisol (nmol/liter)
noradrenaline (nmol/liter)
Adrenaline (nmol/liter)
Reference valuesa 34 - 172 10 - 140 150 - 802 0 - 3.25 0 - 0.55
16h 22h 16h 22h 16h 22h 16h 22h 16h 22h
Patient 1 35 20 41 48 147 159 0.96 0.93 0.18 0.14
Patient 2 71 58 81 71 249 170 1.91 1.03 <0.05b <0.05b
Patient 3 92 49 53 64 480 382 1.23 1.33 0.23 0.22
Patient 4 22 <15b 46 59 232 236 1.16 1.07 0.06 0.09
Patient 5 61 33 54 67 571 334 0.62 0.55 0.09 0.12
Patient 6 62 56 64 76 107 180 0.74 0.51 0.15 0.23
Patient 7 19 <15b 89 90 265 241 1.45 1 0.31 0.41
Patient 8 34 <15b 44 46 524 309 1.26 0.84 0.26 0.14
Patient 9 33 <15b 66 -c 306 130 1.69 1.23 0.21 0.16
Patient 10 39 <15b 58 67 464 323 1.18 0.74 0.15 19
Data are presented as individual values. aReference values for overnight fast. bBelow detection limit. cN = 9.
Table 5 Fat and glucose oxidation
Fat oxidation (mmol/min) Δ (%)
Glucose oxidation (mmol/min) Δ (%)
16h 22h 16h 22h
Patient 1 0.062 0.087 40 0.458 0.197 -57
Patient 2 0.039 0.1 158 1.144 0.425 -63
Patient 3 0.094 0.107 13 0.378 0.051 -87
Patient 4 0.069 0.073 6 0.292 0.126 -57
Patient 5 0.073 0.085 16 0.467 0.264 -43
Patient 6 0.053 0.076 42 0.527 0.307 -42
Patient 7 0.08 0.074 -7 0.212 0.351 65
Patient 8 0.074 0.086 16 0.725 0.45 -38
Patient 10 0.105 0.132 24 0.348 0.000 -100
Data are presented as individual values. N = 9.
proefschrift Soeters.indb 132 4-8-2008 9:13:27
Hypoinsulinem
ic hypoglycemia: a G
aussian Tale
133
Chapter 8Plasma acylcarnitines
Plasma levels of free carnitine as well as short-, medium- and long-chain acylcarnitines
did not exceed upper normal reference limits after 16 h of fasting except for C2-carnitne
in patient 5. Plasma levels of free carnitine as well as short-, medium- and long-chain
acylcarnitines did not exceed upper normal reference limits after 22 h of fasting except
for C2-carnitne in patient 6. Plasma free carnitine decreased in the group as a whole
(median decrease 11 [-12 - 26] %, P = 0.028), whereas the acylcarnitines did not change
for the group between 16 and 22 h of fasting (data not shown).
Organic and amino acid analysisPlasma alanine decreased in most patients between 16 and 22 h of fasting, with a trend
towards a decrease for the total group: median decrease 9 [-22 - 25] %, P = 0.093.
Plasma alanine levels were below reference values in 4 patients both after 16 and 22 h
of fasting. Analysis of other plasma amino acids was uneventful in all patients (data not
shown).
Table 4 Glucoregulatory hormones
Insulin (pmol/liter)
Glucagon (ng/liter)
Cortisol (nmol/liter)
noradrenaline (nmol/liter)
Adrenaline (nmol/liter)
Reference valuesa 34 - 172 10 - 140 150 - 802 0 - 3.25 0 - 0.55
16h 22h 16h 22h 16h 22h 16h 22h 16h 22h
Patient 1 35 20 41 48 147 159 0.96 0.93 0.18 0.14
Patient 2 71 58 81 71 249 170 1.91 1.03 <0.05b <0.05b
Patient 3 92 49 53 64 480 382 1.23 1.33 0.23 0.22
Patient 4 22 <15b 46 59 232 236 1.16 1.07 0.06 0.09
Patient 5 61 33 54 67 571 334 0.62 0.55 0.09 0.12
Patient 6 62 56 64 76 107 180 0.74 0.51 0.15 0.23
Patient 7 19 <15b 89 90 265 241 1.45 1 0.31 0.41
Patient 8 34 <15b 44 46 524 309 1.26 0.84 0.26 0.14
Patient 9 33 <15b 66 -c 306 130 1.69 1.23 0.21 0.16
Patient 10 39 <15b 58 67 464 323 1.18 0.74 0.15 19
Data are presented as individual values. aReference values for overnight fast. bBelow detection limit. cN = 9.
proefschrift Soeters.indb 133 4-8-2008 9:13:27
134
Discussion
The prolonged fast is a well validated clinical test to diagnose and explore spontaneous
hypoglycemia (2). In our experience, some patients display unexplained hypoinsulinemic
(insulin below 42 pmol/L) hypoglycemia (glucose < 3.0) (unpublished observation).
Although it may be reassuring that a diagnosis of insulinoma is excluded in these patients,
a subtle metabolic derangement cannot be ruled out with certainty. In this report we
describe 10 patients who underwent extended metabolic testing to rule out an inborn
error of metabolism, a relatively unexplored area in internal medicine to date.
We investigated a small group of patients which was due to the low numbers of patients
with hypoinsulinemic hypoglycemia but it included patients referred with this problem
over a period of 4 years. From a practical point of view we chose to test patients after
16 h of fasting because our intention was to explore contribution of metabolic defects
resulting in hypoglycemia during the prolonged fasting test and these are biochemically
detectable even in the absence of an overt hypoglycemia (26).
All the patients we studied showed plasma FFA levels within the expected range for
16 h of fasting (27). Plasma FFA levels did not increase in all patients between 16 and
22 h of fasting which is explained by the fact that plasma FFA have been shown not to
increase markedly until 18 to 24 h of fasting (7).
Table 6. Acylcarnitine levels (μmol/liter) after 16 and 22 h of fasting.
Free Carnitine C2-carnitine C4-carnitine C8-carnitine C14:1-carnitine C16:1-carnitine C18:1-carnitine
Reference values 22.3 - 54.8 3.4 - 13.0 0.14 - 0.94 0.04 - 0.22 0.02 - 0.18 0.02 - 0.08 0.06 - 0.28
16h 22h 16h 22h 16h 22h 16h 22h 16h 22h 16h 22h 16h 22h
Patient 1 36.63 31.88 7.25 9.49 0.15 0.2 0.07 0.06 0.09 0.04 0.04 0.04 0.11 0.10
Patient 2 37.29 30.25 2.87 8.58 0.07 0.17 0.07 0.14 0.06 0.14 0.02 0.06 0.13 0.19
Patient 3 34.15 31.16 8.03 9.09 0.28 0.14 0.05 0.04 0.16 0.09 0.07 0.05 0.21 0.13
Patient 4 31.07 34.66 14.56 9.23 0.27 0.12 0.08 0.07 0.08 0.09 0.07 0.04 0.23 0.19
Patient 5 23.97 23.37 12.62 13.32 0.16 0.17 0.04 0.03 0.08 0.06 0.04 0.04 0.13 0.12
Patient 6 33.54 24.86 5.58 8.48 0.19 0.22 0.09 0.08 0.04 0.14 0.03 0.04 0.07 0.1
Patient 7 22.78 19.42 9.94 8.9 0.09 0.06 0.04 0.04 0.06 0.07 0.03 0.03 0.11 0.12
Patient 8 46.49 41.5 6.35 7.23 0.16 0.22 0.09 0.04 0.03 0.07 0.04 0.04 0.12 0.11
Patient 9 25.15 25.25 9.73 9.4 0.27 0.29 0.13 0.09 0.07 0.08 0.03 0.04 0.12 0.1
Patient 10 38.4 30.55 6.52 6.27 0.1 0.09 0.08 0.03 0.19 0.11 0.07 0.05 0.14 0.12
Data are presented as individual values.
proefschrift Soeters.indb 134 4-8-2008 9:13:27
Hypoinsulinem
ic hypoglycemia: a G
aussian Tale
135
Chapter 8
Plasma FFA induce KB formation in the liver (9), a metabolic adaptation that is hampered
in FAO disorders (12). All the patients but one (patient 8) showed KB formation after 22
h of fasting. The lack of KB in patient 8, the only male, may be explained by the plasma
FFA levels that were rather low in comparison with female patients and may not have
stimulated KB formation sufficiently (9). Men have lower plasma FFA levels and lipolysis
rates compared to women during fasting (28).
Plasma free carnitine levels decrease during fasting since carnitine is esterified
intracellular to form acylcarnitine (29): this effect was already observed in our patients
after 22 h of fasting. In general, medium and long-chain acylcarnitines are not increased
after 20 h of fasting in adults (29;30).
MCADD is probably the most frequent occurring FAO disorder in children with incidence
numbers up to 1 in 10,000, but adult presentations have been described (31-33).
Furthermore, the clinical presentation of FAO disorders may vary from asymptomatic
to severe disease, explaining the fact that some cases escaped detection until the
current neonatal screening was employed (31;34;35). Plasma C8-acylcarntine is a main
parameter to diagnose MCADD (12;26;36) and should be higher than 0.3 μmol/liter
for its unequivocal diagnosis (36). In adult MCADD patients plasma C8-carntine levels
can be as high as 5 μmol/liter (26). As plasma C8-carnitine did not exceed 0.14 μmol/
liter despite an increase in FAO, MCADD as a possible explanation for hypoinsulinemic
hypoglycemia was ruled out in our patients. The lack of increased levels of short- and
Table 6. Acylcarnitine levels (μmol/liter) after 16 and 22 h of fasting.
Free Carnitine C2-carnitine C4-carnitine C8-carnitine C14:1-carnitine C16:1-carnitine C18:1-carnitine
Reference values 22.3 - 54.8 3.4 - 13.0 0.14 - 0.94 0.04 - 0.22 0.02 - 0.18 0.02 - 0.08 0.06 - 0.28
16h 22h 16h 22h 16h 22h 16h 22h 16h 22h 16h 22h 16h 22h
Patient 1 36.63 31.88 7.25 9.49 0.15 0.2 0.07 0.06 0.09 0.04 0.04 0.04 0.11 0.10
Patient 2 37.29 30.25 2.87 8.58 0.07 0.17 0.07 0.14 0.06 0.14 0.02 0.06 0.13 0.19
Patient 3 34.15 31.16 8.03 9.09 0.28 0.14 0.05 0.04 0.16 0.09 0.07 0.05 0.21 0.13
Patient 4 31.07 34.66 14.56 9.23 0.27 0.12 0.08 0.07 0.08 0.09 0.07 0.04 0.23 0.19
Patient 5 23.97 23.37 12.62 13.32 0.16 0.17 0.04 0.03 0.08 0.06 0.04 0.04 0.13 0.12
Patient 6 33.54 24.86 5.58 8.48 0.19 0.22 0.09 0.08 0.04 0.14 0.03 0.04 0.07 0.1
Patient 7 22.78 19.42 9.94 8.9 0.09 0.06 0.04 0.04 0.06 0.07 0.03 0.03 0.11 0.12
Patient 8 46.49 41.5 6.35 7.23 0.16 0.22 0.09 0.04 0.03 0.07 0.04 0.04 0.12 0.11
Patient 9 25.15 25.25 9.73 9.4 0.27 0.29 0.13 0.09 0.07 0.08 0.03 0.04 0.12 0.1
Patient 10 38.4 30.55 6.52 6.27 0.1 0.09 0.08 0.03 0.19 0.11 0.07 0.05 0.14 0.12
Data are presented as individual values.
proefschrift Soeters.indb 135 4-8-2008 9:13:27
136
long-chain acylcarnitines in our patients negates other defects of FAO (i.e. short- and
long-chain acyl-CoA dehydrogenase deficiency) to be present.
Fasting increases fat oxidation and decreases glucose oxidation. These changes are
present within 24 h of fasting (7). A normal increase of whole body FAO was observed in
9 patients between 16 and 22 h of fasting. Patient 7 showed no increase of FAO which
is not readily explained since no other abnormalities were found.
Defective gluconeogenesis and glycogenolysis may be accompanied by elevated
gluconeogenic amino acids and in increase of various metabolites such as lactate
and pyruvate (37). Alanine is a critical gluconeogenic amino acid that is converted
into pyruvate, notably during catabolic stress (12;38). Plasma alanine levels decrease
during the first 30 h of fasting (39). The lower plasma alanine concentrations argue
against defective EGP. Additionally, no abnormalities in other amino acids or organic
were detected, excluding amino/organic acid disorders as a cause for hypoinsulinemic
hypoglycemia in these patients.
Plasma insulin levels were low in all patients as expected (7;8). The other glucoregulatory
hormones showed no abnormalities, although these hormones have a circadian or
pulsatile rhythm so minor individual changes may not have been discovered (8).
In this report we described 10 patients who underwent extended metabolic testing
to rule out decreased EGP in possible combination with FAO disorders like MCADD and
amino/organic acid disorders. We found normal plasma glucose concentrations and
glucose production without disorders in gluconeogenic amino/organic acids whereas
plasma acylcarnitine profiling showed no FAO disorders. These data thus show that
hypoinsulinemic hypoglycemia in these adults is not caused by an inborn error of
metabolism. The decrease in plasma glucose concentrations and EGP between 16 and
22 h of fasting seem perfectly in agreement with the earlier reported decrease of glucose
concentrations and EGP between 16 and 22 h of fasting in lean healthy men and women
(27;40). The same holds true for glucose clearance. These data suggest that plasma
glucose values may decrease below the “classical” threshold value for hypoglycemia during
fasting and therefore could represent simply the lower tail of the Gaussian Curve.
AcknowledgementsThe authors wish to thank A.F.C Ruiter for excellent assistance on biochemical and stable
isotope analyses.
proefschrift Soeters.indb 136 4-8-2008 9:13:27
Hypoinsulinem
ic hypoglycemia: a G
aussian Tale
137
Chapter 8
Reference List 1. Service FJ. Hypoglycemic Disorders. N Engl J Med 1995; 332(17):1144-1152.
2. Service FJ, Natt N. The Prolonged Fast. J Clin Endocrinol Metab 2000; 85(11):3973-3974.
3. Boyle PJ, Shah SD, Cryer PE. Insulin, glucagon, and catecholamines in prevention of hypoglycemia during fasting. Am J Physiol Endocrinol Metab 1989; 256(5):E651-E661.
4. Soeters MR, Sauerwein HP, Dubbelhuis PF et al. Muscle adaptation to short-term fasting in healthy lean humans. J Clin Endocrinol Metab 2008;jc.
5. Clore JN, Glickman PS, Helm ST, Nestler JE, Blackard WG. Accelerated decline in hepatic glucose production during fasting in normal women compared with men. Metabolism 1989; 38(11):1103-1107.
6. Merimee TJ, Fineberg SE. Homeostasis during fasting. II. Hormone substrate differences between men and women. J Clin Endocrinol Metab 1973; 37(5):698-702.
7. Klein S, Sakurai Y, Romijn JA, Carroll RM. Progressive alterations in lipid and glucose metabolism during short-term fasting in young adult men. Am J Physiol Endocrinol Metab 1993; 265(5):E801-E806.
8. Hojlund K, Wildner-Christensen M, Eshoj O et al. Reference intervals for glucose, {beta}-cell polypeptides, and counterregulatory factors during prolonged fasting. Am J Physiol Endocrinol Metab 2001; 280(1):E50-E58.
9. Fukao T, Lopaschuk GD, Mitchell GA. Pathways and control of ketone body metabolism: on the fringe of lipid biochemistry. Prostaglandins, Leukotrienes and Essential Fatty Acids 2004; 70(3):243-251.
10. Fery F. Role of hepatic glucose production and glucose uptake in the pathogenesis of fasting hyperglycemia in type 2 diabetes: normalization of glucose kinetics by short-term fasting. J Clin Endocrinol Metab 1994; 78(3):536-542.
11. Wiesli P, Brandle M, Schwegler B, Lehmann R, Spinas GA, Schmid C. A plasma glucose concentration below 2.5 mmol L-1 is not an appropriate criterion to end the 72-h fast. Journal of Internal Medicine 2002; 252(6):504-509.
12. Olpin SE. Implications of impaired ketogenesis in fatty acid oxidation disorders. Prostaglandins, Leukotrienes and Essential Fatty Acids 2004; 70(3):293-308.
13. Iafolla AK, Thompson Jr RJ, Roe CR. Medium-chain acyl-coenzyme A dehydrogenase deficiency: Clinical course in 120 affected children. The Journal of Pediatrics 1994; 124(3):409-415.
14. Wanders RJ, Vreken P, den Boer ME, Wijburg FA, van Gennip AH, IJlst L. Disorders of mitochondrial fatty acyl-CoA beta-oxidation. J Inherit Metab Dis 1999; 22(4):442-487.
15. Endert E, Ouwehand A, Fliers E, Prummel MF, Wiersinga WM. Establishment of reference values for endocrine tests. Part IV: Adrenal insufficiency. Neth J Med 2005; 63(11):435-443.
16. Ackermans MT, Pereira Arias AM, Bisschop PH, Endert E, Sauerwein HP, Romijn JA. The quantification of gluconeogenesis in healthy men by (2)H2O and [2-(13)C]glycerol yields different results: rates of gluconeogenesis in healthy men measured with (2)H2O are higher than those measured with [2-(13)C]glycerol. J Clin Endocrinol Metab 2001; 86(5):2220-2226.
17. van Vlies N, Tian L, Overmars H et al. Characterization of carnitine and fatty acid metabolism in the long-chain acyl-CoA dehydrogenase-deficient mouse. Biochem J 2005; 387(1):185-193.
18. Langeveld M, Ghauharali KJM, Sauerwein HP et al. Type I Gaucher disease, a glycosphingolipid storage disorder, is associated with insulin resistance. J Clin Endocrinol Metab 2007;jc.
19. de Metz J, Sprangers F, Endert E et al. Interferon-gamma has immunomodulatory effects with minor endocrine and metabolic effects in humans. J Appl Physiol 1999; 86(2):517-522.
20. Sprangers F, Jellema WT, Christa E et al. Partial inhibition of nitric oxide synthesis in vivo does not inhibit glucose production in man. Metabolism 2002; 51(1):57-64.
proefschrift Soeters.indb 137 4-8-2008 9:13:27
138
21. Birjmohun RS, Bisoendial RJ, van Leuven SI et al. A single bolus infusion of C-reactive protein increases gluconeogenesis and plasma glucose concentration in humans. Metabolism 2007; 56(11):1576-1582.
22. Allick G, van der Crabben SN, Ackermans MT, Endert E, Sauerwein HP. Measurement of gluconeogenesis by deuterated water: the effect of equilibration time and fasting period. Am J Physiol Endocrinol Metab 2006; 290(6):E1212-E1217.
23. Bisschop PH, Arias AMP, Ackermans MT et al. The Effects of Carbohydrate Variation in Isocaloric Diets on Glycogenolysis and Gluconeogenesis in Healthy Men. J Clin Endocrinol Metab 2000; 85(5):1963-1967.
24. Frayn KN. Calculation of substrate oxidation rates in vivo from gaseous exchange. J Appl Physiol 1983; 55(2):628-634.
25. van der Crabben SN, Allick G, Ackermans MT, Endert E, Romijn JA, Sauerwein HP. Prolonged Fasting Induces Peripheral Insulin Resistance, Which Is Not Ameliorated by High-Dose Salicylate. J Clin Endocrinol Metab 2008; 93(2):638-641.
26. Huidekoper H, Schneider J, Westphal T, Vaz F, Duran M, Wijburg F. Prolonged moderate-intensity exercise without and with L -carnitine supplementation in patients with MCAD deficiency. Journal of Inherited Metabolic Disease 2006; 29(5):631-636.
27. Corssmit EPM, Stouthard JML, Romijn JA, Endert E, Sauerwein HP. Sex differences in the adaptation of glucose metabolism to short-term fasting: Effects of oral contraceptives. Metabolism 1994; 43(12):1503-1508.
28. Soeters MR, Sauerwein HP, Groener JE et al. Gender-Related Differences in the Metabolic Response to Fasting. J Clin Endocrinol Metab 2007; 92(9):3646-3652.
29. Hoppel CL, Genuth SM. Carnitine metabolism in normal-weight and obese human subjects during fasting. Am J Physiol 1980; 238(5):E409-E415.
30. Costa CCG, De Almeida IT, Jakobs Corn , Poll-The B-T, Duran M. Dynamic Changes of Plasma Acylcarnitine Levels Induced by Fasting and Sunflower Oil Challenge Test in Children. [Miscellaneous Article]. Pediatric Research 1999; 46(4):440.
31. Grosse SD, Khoury MJ, Greene CL, Crider KS, Pollitt RJ. The epidemiology of medium chain acyl-CoA dehydrogenase deficiency: an update. Genet Med 2006; 8(4):205-212.
32. Bodman M, Smith D, Nyhan WL, Naviaux RK. Medium-Chain Acyl Coenzyme A Dehydrogenase Deficiency: Occurrence in an Infant and His Father. Arch Neurol 2001; 58(5):811-814.
33. Feillet F, Steinmann G, Vianey-Saban C et al. Adult presentation of MCAD deficiency revealed by coma and severe arrythmias. Intensive Care Medicine 2003; 29(9):1594-1597.
34. Fromenty B, Mansouri A, Bonnefont JP et al. Most cases of medium-chain acyl-CoA dehydrogenase deficiency escape detection in France. Hum Genet 1996; 97(3):367-368.
35. Roe CR, Millington DS, Maltby DA, Kinnebrew P. Recognition of medium-chain acyl-CoA dehydrogenase deficiency in asymptomatic siblings of children dying of sudden infant death or Reye-like syndromes. J Pediatr 1986; 108(1):13-18.
36. Van Hove JL, Zhang W, Kahler SG et al. Medium-chain acyl-CoA dehydrogenase (MCAD) deficiency: diagnosis by acylcarnitine analysis in blood. Am J Hum Genet 1993; 52(5):958-966.
37. Saudubray JM, Sedel F, Walter JH. Clinical approach to treatable inborn metabolic diseases: an introduction. J Inherit Metab Dis 2006; 29(2-3):261-274.
38. van den Berghe G. Disorders of gluconeogenesis. Journal of Inherited Metabolic Disease 1996; 19(4):470-477.
39. Haymond MW, Karl IE, Clarke WL, Pagliara AS, Santiago JV. Differences in circulating gluconeogenic substrates during short-term fasting in men, women, and children. Metabolism 1982; 31(1):33-42.
40. Romijn JA, Godfried MH, Hommes MJT, Endert E, Sauerwein HP. Decreased glucose oxidation during short-term starvation. Metabolism 1990; 39(5):525-530.
proefschrift Soeters.indb 138 4-8-2008 9:13:27
proefschrift Soeters.indb 139 4-8-2008 9:13:27
proefschrift Soeters.indb 140 4-8-2008 9:13:28
The Metabolic Response to Fasting in Humans: a Perspective.9
proefschrift Soeters.indb 141 4-8-2008 9:13:28
proefschrift Soeters.indb 142 4-8-2008 9:13:28
The Metabolic Response to Fasting in H
umans: a Perspective.
143
Chapter 9
Introduction
The metabolic response to fasting consists of integrated adaptations that prevent energy
depletion. Such adaptations protect the organism from hypoglycemia and protein losses
to increase the chances of survival during starvation (1-3). Short-term fasting can be
defined as the first 3 days of starvation in which progressive alterations in lipid and
glucose metabolism occur (4-9). Although this definition may be arbitrary and debated,
it is based on earlier observations that the adaptation to fasting is more or less maximal
within about three days without further changes afterwards.
Fasting has been studied as a model of semi-acute lipid exposure (8;10-14), but also,
perhaps as a consequence, as a model to study the interference of lipids with glucose
metabolism, with special reference to the decrease of insulin mediated glucose uptake
(4;15-21).
In the perspective of this thesis, the metabolic muscle adaptation to short-term fasting
is discussed for its appropriateness as model for free fatty acid (FFA) induced insulin
resistance (chapter 2, 3 and 4). In addition, turnover and oxidation of substrates like
lipid and glucose in short-term fasting are discussed in relation to several mechanisms
(chapter 2, 3 and 4). Although the thrifty gene concept has been debated (22-24), the
adaptation to intermittent fasting is addressed since it may provide in a simple tool to
increase peripheral insulin sensitivity (25) (chapter 7). Then, the adaptive changes in
ketone body (KB) metabolism are outlined, with special reference to a newly identified
alternative KB pathway (chapter 5 and 6). Finally, directions for future research on the
metabolic response to fasting in humans are contemplated.
The Metabolic Response to Fasting: free fatty acid induced insulin resistance?
The only situation that causes “natural” insulin resistance in healthy lean subjects is
starvation (26). Earlier studies showed decreased insulin sensitivity (i.e. decreased
peripheral insulin mediated glucose uptake) after short-term fasting (4;15-21). The
decreased glucose disposal was poorly explained by the idea that fasting induced a
metabolic milieu favoring glucose production and the use of other fuels than glucose:
it would then take a considerable time after refeeding to reverse this adaptation
(4;16;17;19;21). More recently it has been tried to elucidate the fasting induced peripheral
insulin resistance by suggesting that the increased plasma FFA and intramyocellular
proefschrift Soeters.indb 143 4-8-2008 9:13:28
144
lipid (IMCL) during short-term fasting may interfere with peripheral insulin sensitivity
(15;26;27), analogous to the proposed accumulation of lipid metabolites in obesity
induced insulin resistance (OIIR) and type 2 diabetes mellitus (DM) (28;29). High plasma
FFA are suggested to interfere with peripheral insulin mediated glucose uptake, but the
exact mechanisms are still not fully elucidated (30;31). Interestingly women are protected
from FFA induced insulin resistance compared to matched men (32;33).
Various metabolites of FFA have been suggested to decrease insulin sensitivity in
skeletal muscle (34). The sphingolipid ceramide is one of these mediators. Increased
muscle ceramide concentrations have been reported in skeletal muscle of obese insulin
resistant subjects (28;35), while a negative correlation of muscle ceramide with insulin
sensitivity was found (35). However, skeletal muscle ceramide levels did not seem to play
an important role in FFA associated insulin resistance in other studies (36;37).
Intramyocellular ceramide content is mainly dependent on de novo synthesis from fatty
acids (38). In vitro studies showed that intracellular ceramide synthesis from palmitate is
one of the mechanisms by which palmitate interferes negatively with insulin-stimulated
phosphorylation of protein kinase B/AKT (AKT) (39;40). Ceramide was shown to increase
in rodents during fasting (41).
Therefore we examined the metabolic adaptation to fasting in relation to muscle
ceramide levels. We found that after fasting for 38 h in lean healthy men and women
no correlation exists between muscle ceramide and insulin sensitivity (42). Despite higher
plasma FFA levels in the basal state, women were equally insulin-sensitive compared to
men, suggesting a relative protection from FFA-induced insulin resistance in women as
in other reports (32;33). To detect potential differences in FFA-uptake and fatty acid
handling we assessed muscle concentrations of ceramide and the fatty acid transporter
(FAT)/CD36 during the clamp. The muscle ceramide levels did not differ between females
and males. Additionally, there were no differences in total FAT/CD36 amount between
women and men in muscle, which is remarkable since it has been described earlier that
women have higher levels of this lipid-binding protein (in muscle) after an overnight fast
(8). Our data suggest that the initial gender differences in intramyocellular lipids (16)
and fatty acid transporters after an overnight fast (8) are abolished after 38 h of fasting.
We concluded that women are less sensitive to FFA-induced insulin resistance which is
not explained by differences in muscle ceramide or FAT/CD36. Since we did find higher
lipolysis rates in women without increased FAO, it is puzzling how the plasma FFA in
women are disposed. Uptake of plasma FFA predominantly occurs in liver, adipocytes
and muscle. FFA recycling occurs in the post-absorptive state and women display greater
efficiency in direct FFA uptake in subcutaneous tissue (femoral area) (43). The liver may
proefschrift Soeters.indb 144 4-8-2008 9:13:28
The Metabolic Response to Fasting in H
umans: a Perspective.
145
Chapter 9
be another site of increased non-oxidative FFA disposal. Although livers of females do not
contain more fat than male livers (44), it is known that female livers contain more FAT/
CD36 and have a greater capacity for FFA uptake and synthesis of ketone bodies and
very low-density lipoproteins-triglycerides (45-47).
Since gender seems to fundamentally influence the lipid-glucose interplay (13;47-49)
future studies are needed to explain how women are relatively protected from FFA-
induced insulin resistance. This mechanistic insight could provide us a fundamental basis
to develop strategies to reduce obesity-induced insulin resistance. Moreover these gender
differences show that men and women should be examined separately in these studies.
Because the above mentioned study did not investigate the change in muscle ceramide
in time during fasting, we investigated lean healthy men after 14 and 62 h of fasting
focusing on muscle ceramide and the insulin signaling cascade because muscle ceramide
has been shown to induce insulin resistance by interfering directly with the insulin signaling
cascade (38) with most of the attention focusing on AKT (38;40;50-54). AKT is a 56 kD
serine/threonine kinase and a central mediator of many insulin effects with complex
regulation (52). Ceramide has been shown to interfere with the two phosphorylation
steps on the serine473- and threonine308-residues of AKT (38;52;52-54).
We found that fasting for 62 h tends to increase muscle ceramide in young healthy
volunteers though no correlation between muscle ceramide and insulin sensitivity was
found (55). However we did find a trend towards lower pAKT-ser473 and a significant
decrease of the ratio pAKT-ser473/tAKT after 62 h of fasting. This is in contrast with
the study of Bergman et al, who reported no differences of AKT between 12 and 48
h of fasting. Whether the difference in fasting duration of the two studies may explain
these differences is not known at present. Whether the decrease in pAKT-ser473 during
hyperinsulinemia after 62 hours of fasting can be attributed to the trend to the higher
muscle ceramide levels in the basal state remains unanswered, since we found no
correlation of ceramide with peripheral insulin sensitivity in line with previous observations
(36;37;42). Since studies in OIIR and type 2 DM have shown effects on pAKT-thr308 (56)
but not on pAKT-ser473 (56;57), it is possible that pAKT-ser473 is only involved in the
physiological adaptation to fasting, inducing a reduction in peripheral glucose uptake
and thereby protecting the body from hypoglycemia. Further studies are needed no
investigate upstream mediators of AKT.
To be oxidized, activated long-chain fatty acids can only transverse the mitochondrial
membranes as acylcarnitines (ACs, an activated long-chain fatty acid that is coupled to
carnitine) (58). Recently however, muscle ACs have been implicated in insulin resistance
proefschrift Soeters.indb 145 4-8-2008 9:13:28
146
via a currently unknown mechanism (59-61). These studies suggested that increased beta-
oxidation disrupts the tricarboxylic acid cycle (TCA) with subsequent inhibition of complete
fatty acid oxidation via a high energy redox state (rising NADH/NAD+ and acetyl-CoA/free
CoA ratios). The accumulation of metabolic by-products (e.g. acylcarnitines) would then
activate stress kinases or other signals, interfering with insulin action (60). However, how
and whether ACs directly affect insulin mediated glucose uptake is currently unraveled.
To explore the acylcarnitine profiles in skeletal muscle and their possible role in the
induction of insulin resistance, we investigated whether fasting elevates muscle long-chain
ACs, since this is true for plasma long-chain ACs in humans (62-64), and whether this
would correlate with peripheral insulin sensitivity. Such an increase of ACs during fasting
could match increased lipid oxidation and possibly decreased peripheral insulin sensitivity
(6;15;55). However, we demonstrated that muscle long-chain ACs do not increase
during short-term fasting despite an increase in whole body fatty acid oxidation (FAO).
Surprisingly, under hyperinsulinemic circumstances muscle long-chain ACs decreased after
14 h of fasting, but not after 62 h of fasting. However we could not detect a correlation
of muscle long-chain ACs and peripheral insulin sensitivity. Besides the consideration that
the muscle concentration of long-chain ACs during fasting in humans does not reflect
the fatty acid oxidation flux, muscle long-chain ACs are unlikely to play a causal role in
peripheral insulin resistance during fasting.
The Metabolic Response to Fasting: tenacious changes in substrate oxidation?
During short-term fasting, FAO increases with a concomitant decrease of glucose
oxidation (CHO) (7;65). Fasting increases the intramyocellular AMP/ATP ratio that
activates AMP-activated protein kinase (AMPK) by AMP binding and phosphorylation
(66). Phosphorylated AMPK then phosphorylates and inhibits ACC activity, thereby
inhibiting malonyl-CoA synthesis. This results in decreased inhibition of CPT-1 activity,
thereby increasing mitochondrial import and FAO in muscle. The increase in FAO will yield
acetyl-CoA which inhibits glycolysis and subsequent glucose oxidation via the pyruvate
dehydrogenase complex (PDHc). These changes of FAO and CHO may be an example of
Randle’s glucose - fatty acid cycle (67). FAO yields a substantial amount of acetyl-CoA
that accumulates in order to be oxidized. The accumulated acetyl-CoA increases pyruvate
dehydrogenase kinase (PDK4) that results in decreased pyruvate dehydrogenase (PDH)
thereby lowering glucose oxidation (65). In contrast, the Randle cycle was revisited by
proefschrift Soeters.indb 146 4-8-2008 9:13:28
The Metabolic Response to Fasting in H
umans: a Perspective.
147
Chapter 9
Boden and Shulman by argumenting that in peripheral insulin resistance, the decreased
transmembrane glucose transport via GLUT 4 is rate limiting and not accumulation of
glucose-6-phosphate as originally suggested by Randle (68). How this reduction in GLUT
4 translocation to the plasma membrane occurs is still matter of debate and probably
multifactorial in a setting of lipid overload.
The metabolic changes that occur after short-term fasting during hyperinsulinemic
conditions are of interest since these reflect the ability of the organism to change substrate
oxidation in response to insulin. The hyperinsulinemic euglycemic clamp increases CHO,
and it has been shown that the CHO during hyperinsulinemia is lower after short-term
fasting compared to an overnight fast (15;69). More over it was suggested that the
hyperinsulinemia induced increase in CHO was lower after short-term fasting (15;69),
which is called metabolic inflexibility.
However, we show that the increase in CHO during the hyperinsulinemic euglycemic
clamp is not different after short-term fasting compared to an overnight fast although
the point of departure is lower after short-term fasting. Also, non oxidative glucose
disposal (NOGD) expressed as percentage of insulin mediated glucose uptake is not
different before and after short-term fasting. NOGD expressed as part of glucose uptake
takes into account the changes in glucose uptake as reviewed by Boden (68). These
data are strengthened by our finding that phosphorylation of glycogen synthase kinase
(GSK)-3 was not lower during hyperinsulinemia after short-term fasting compared with
an overnight fast. Phosphorylation of GSK by AKT stimulates glycogen synthesis (20).
Analogous to the changes in CHO, the decrease in FAO during hyperinsulinemia
is not different after short-term fasting compared to an overnight fast although the
point of departure now is higher after short-term fasting. This argues with the previous
suggestion that fasting induces metabolic inflexibility (15). These findings bring up the
question whether metabolic inflexibility as defined by a switch in substrate oxidation
upon stimulation truly exists and is not just a matter of expression of the data as absolute
flux rates instead of taking the basal levels into account.
proefschrift Soeters.indb 147 4-8-2008 9:13:28
148
The Metabolic Response to Fasting: a newly identified alternative pathway in ketone body metabolism
Besides increased lipid oxidation, the KBs, D-3-hydroxybutyrate (D-3HB) and acetoacetate
(AcAc), are an important fuel during fasting (70;71). Ketogenesis in lean healthy subjects
only takes place during fasting when fatty acids reach the liver and are degraded within
liver mitochondria. Acetyl-CoA then is channeled into the ketogenesis pathway (70).
The transport of newly synthesized KBs from the hepatocytes to the circulation and the
subsequent uptake in extrahepatic tissues occurs via diffusion and transporters (71;72).
It has been suggested by at least two separate studies that 3-hydroxybutyrylcarnitine
(3HB-carnitine) is the product of the coupling of D-3HB and carnitine (61;73). Although it
has been described that D-3HB can be activated to D-3HB-CoA in rat liver and hepatoma
cells, this has not been established in humans (74-76). Furthermore, it is unknown
whether and how D-3HB-CoA can be coupled to carnitine resulting in D-3HB-carnitine.
This is in contrast to its generally known stereo isomer L-hydroxybutyryl-CoA (L-3HB-
CoA), formed via beta-oxidation of butyryl-CoA (77). Although it was proposed that the
total amount of tissue hydroxybutyrylcarnitine was derived from D-3HB and carnitine, the
quantative contributions of the L and D stereo isomers were not accounted for in these
studies (61;73).
In one of the studies presented in this thesis, we found a 32 fold increase of muscle
3HB-carnitine after 62 h of fasting (unpublished data). Therefore we explored KB
metabolism in relation to the different stereo isomers of 3HB-carnitine. We found that
the bulk (88%) of muscle 3HB-carnitine during fasting in lean healthy men consists of
D-3HB-carnitine (78). Moreover muscle D-3HB-carnitine correlates well with the turnover
of D-3HB. In addition, we confirmed that succinyl 3-ketoacyl-CoA transferase (SCOT) may
be the responsible enzyme for the activation of D-3HB in muscle and liver homogenate
studies (78). Subsequent exchange of CoA for carnitine then would be the following step
and we have shown that this likely occurs via CAT (or another acylcarnitine transferase
like CPT2) (79).
Currently it is unknown why D-3HB is activated and subsequently coupled to carnitine.
Several explanations might be plausible: 1. reflection of true limitation of KB oxidation
with the coupling to carnitine preventing mitochondrial accumulation (79;80), 2.
transport needs between compartments and a reservoir of energy (79) or 3. preventing
ketoacidosis during fasting. Finally D-3HB may be used for lipogenesis in muscle as it is
proefschrift Soeters.indb 148 4-8-2008 9:13:28
The Metabolic Response to Fasting in H
umans: a Perspective.
149
Chapter 9
in liver (75;76). This would be in agreement with the increase in IMCL during fasting
(27;81;82).
The Metabolic Response to Intermittent Fasting: always thrifty and beneficial?
Intermittent fasting (IF) is characterized by refraining form food intake in certain
intervals and was studied because it was postulated to have beneficial effects on energy
metabolism in light of the thrifty gene concept (25). The “thrifty” genotype was defined
as “being exceptionally efficient in the intake and/or utilization of food” by Neel (24).
Animal and human studies have shown effects of IF on energy metabolism (83). But the
latter have not been numerous and did not yield equivocal data (25;84;85). However, the
increase of insulin sensitivity after two weeks of IF shown by Halberg et al is of interest
since it may provide in a simple tool to improve insulin sensitivity (25).
IF consists of repetitive bouts of short-term fasting that may alter peripheral insulin
sensitivity as well as hepatic insulin sensitivity (25) and proteolysis (86;87), although
protein stores are protected from excessive degradation (2;88). We found that
intermittent fasting only increases peripheral insulin sensitivity with regard to glucose
uptake and not hepatic insulin sensitivity after two weeks of eucaloric intermittent
fasting. This effect is overcome by prolonged hyperinsulinemia. Also IF increased the
insulin mediated inhibition of proteolysis. Insulin sensitivity of adipose tissue tends to be
augmented where little effect was observed on proteolysis. More over resting energy
expenditure (REE) decreased after two weeks of intermittent fasting.
Besides this effect on whole body kinetics, we showed that IF alters insulin and
mammalian target of rapamycine (mTOR) signaling. We found a trend towards higher
phosphorylation of AKT after IF and higher phosphorylation of GSK in the basal state
and clamp. Such adaptive response may aid in replenishing glycogen levels. Additionally
mTOR phosphorylation was decreased, suggesting that the inhibition of proteolysis is
accompanied by a decrease in protein synthesis (89). Whether IF would affect skeletal
muscle mass on the long run remains elusive.
These metabolic adaptations are of interest since they show that the body adapts to
this situation in a thrifty way. However, the data also imply that intermittent fasting (e.g.
“De Balansdag” in the Netherlands) may be disadvantageous because the decrease in REE
may eventually lead to an increase in weight even under seemingly eucaloric conditions.
To what extend IF can be beneficial in obese and diabetic patients remains elusive. The
proefschrift Soeters.indb 149 4-8-2008 9:13:28
150
mechanism that decreases REE during intermittent fasting is of great importance in the
design of future studies on this matter.
The Metabolic Response to Fasting: What is the purpose?
The answer to this question may be simple: to safe energy. Although some of the
adaptations to fasting may be known, the exact reason for these adaptations is unknown
and may only be understood once refeeding is studied.
Increased lipolysis consequently increases plasma FFA, but also intracellular lipids on
distant sites in both liver and muscle (27;81;90). In muscle this reesterification of FFA to
yield intramyocellular triglycerides (IMCL) exceeds the rate of simultaneous intracellular
triglyceride fatty acid oxidation (82). The synthesis of D-3HB-carnitine, as shown in this
thesis, may be a comparable phenomenon.
The reasons for this discrepancy between muscle FAO and lipid supply during short-
term fasting remain to be elucidated, though Stannard and Johnson speculated that
storage of lipid would afford a survival advantage during periods of starvation by ensuring
adequate fuel for necessary muscle function and protecting remaining blood glucose for
the brain (26). The IMCL function as fuel for muscle during fasting may be challenged.
Muscle glycogen stores are not broken into during 84 of fasting and remain available
for activity opposed to liver glycogen that is depleted after 40 h of fasting (91;92). Also
exercise during fasting induced higher levels of plasma FFA and KB compared to exercise
after an overnight fast indicating that increased fuel supply is needed for activity (91).
Data on IMCL during exercise after short-term fasting are lacking. IMCL also seem to
be a marker for lipid overload in states of insulin resistance, i.e. obesity (93). Therefore,
perhaps the discrepancy between muscle FAO and lipid supply during fasting plays a role
in peripheral insulin resistance.
Decreased insulin sensitivity during fasting is highly regulated as shown in this thesis
(55) and possibly mediated via FFA metabolites in muscle (15;26;27;55;81). It has been
suggested by others, and ourselves, that this insulin resistance may prevent hypoglycemia.
This may be a feeble explanation. During fasting insulin levels are typically very low with
reciprocal changes in other glucoregulatory hormones (94) inducing EGP, lipolysis and
the adaptations in substrate oxidation. Decreased insulin sensitivity makes sense in the
presence of plasma insulin levels that stimulate glucose uptake via the insulin signaling
cascade translocating the insulin specific glucose transporter GLUT4 (95). Decreased
proefschrift Soeters.indb 150 4-8-2008 9:13:28
The Metabolic Response to Fasting in H
umans: a Perspective.
151
Chapter 9
insulin sensitivity may play a role during refeeding for two reasons. First, it may prevent
massive uptake of glucose thereby preventing refeeding hypoglycemia. The other option
is that decreased insulin sensitivity during refeeding lowers accumulated IMCL by, as we
showed, ongoing FAO. Such mechanism could prevent the noxious effects of chronic
lipid accumulation. Although it is the believe of the author that the latter mechanism is
present, studies should be aimed on how the increased IMCL are lowered again during
refeeding after short-term fasting and what biochemical pathways are responsible. Such
knowledge may aid to increase peripheral insulin sensitivity in OIIR and type 2 DM.
As discussed, other area’s of interest include the gender difference in lipid induced
peripheral insulin resistance, detailed effects of fasting on insulin signaling and peripheral
glucose uptake, D-3HB-carnitine and its routes of composition and degradation. It should
be stressed in a thesis with studies on human physiology that one cannot comprehend
the pathophysiology of insulin resistance, when the physiological sense of the metabolic
adaptations to fasting is not understood.
Reference List 1. Cahill GF, Jr., Herrera MG, Morgan AP et al. Hormone-fuel interrelationships during fasting. J
Clin Invest 1966; 45(11):1751-1769.
2. Cahill J. Starvation in man. Clinics in Endocrinology and Metabolism 1976; 5(2):397-415.
3. Owen OE, Morgan AP, Kemp HG, Sullivan JM, Herrera MG, Cahill GF, Jr. Brain metabolism during fasting. J Clin Invest 1967; 46(10):1589-1595.
4. Bjorkman O, Eriksson LS. Influence of a 60-hour fast on insulin-mediated splanchnic and peripheral glucose metabolism in humans. J Clin Invest 1985; 76(1):87-92.
5. Clore JN, Glickman PS, Helm ST, Nestler JE, Blackard WG. Accelerated decline in hepatic glucose production during fasting in normal women compared with men. Metabolism 1989; 38(11):1103-1107.
6. Klein S, Sakurai Y, Romijn JA, Carroll RM. Progressive alterations in lipid and glucose metabolism during short-term fasting in young adult men. Am J Physiol Endocrinol Metab 1993; 265(5):E801-E806.
7. Carlson MG, Snead WL, Campbell PJ. Fuel and energy metabolism in fasting humans. Am J Clin Nutr 1994; 60(1):29-36.
8. Klein S, Young VR, Blackburn GL, Bistrian BR, Wolfe RR. Palmitate and glycerol kinetics during brief starvation in normal weight young adult and elderly subjects. J Clin Invest 1986; 78(4):928-933.
9. Haymond MW, Karl IE, Clarke WL, Pagliara AS, Santiago JV. Differences in circulating gluconeogenic substrates during short-term fasting in men, women, and children. Metabolism 1982; 31(1):33-42.
10. Horowitz JF, Coppack SW, Paramore D, Cryer PE, Zhao G, Klein S. Effect of short-term fasting on lipid kinetics in lean and obese women. Am J Physiol Endocrinol Metab 1999; 276(2):E278-E284.
11. Klein S, Peters EJ, Holland OB, Wolfe RR. Effect of short- and long-term beta-adrenergic blockade on lipolysis during fasting in humans. Am J Physiol 1989; 257(1 Pt 1):E65-E73.
proefschrift Soeters.indb 151 4-8-2008 9:13:28
152
12. Klein S, Holland OB, Wolfe RR. Importance of blood glucose concentration in regulating lipolysis during fasting in humans. Am J Physiol 1990; 258(1 Pt 1):E32-E39.
13. Mittendorfer B, Horowitz JF, Klein S. Gender differences in lipid and glucose kinetics during short-term fasting. Am J Physiol Endocrinol Metab 2001; 281(6):E1333-E1339.
14. Wolfe RR, Peters EJ, Klein S, Holland OB, Rosenblatt J, Gary H, Jr. Effect of short-term fasting on lipolytic responsiveness in normal and obese human subjects. Am J Physiol Endocrinol Metab 1987; 252(2):E189-E196.
15. Bergman BC, Cornier MA, Horton TJ, Bessesen DH. Effects of fasting on insulin action and glucose kinetics in lean and obese men and women. Am J Physiol Endocrinol Metab 2007; 293(4):E1103-E1111.
16. Webber J, Simpson E, Parkin H, Macdonald IA. Metabolic effects of acute hyperketonaemia in man before and during an hyperinsulinaemic euglycaemic clamp. Clin Sci (Lond) 1994; 86(6):677-687.
17. Newman WP, Brodows RG. Insulin action during acute starvation: evidence for selective insulin resistance in normal man. Metabolism 1983; 32(6):590-596.
18. van der Crabben SN, Allick G, Ackermans MT, Endert E, Romijn JA, Sauerwein HP. Prolonged Fasting Induces Peripheral Insulin Resistance, Which Is Not Ameliorated by High-Dose Salicylate. J Clin Endocrinol Metab 2008; 93(2):638-641.
19. Mansell PI, Macdonald IA. The effect of starvation on insulin-induced glucose disposal and thermogenesis in humans. Metabolism 1990; 39(5):502-510.
20. Anderson JW, Herman RH. Effects of carbohydrate restriction on glucose tolerance of normal men and reactive hypoglycemic patients. Am J Clin Nutr 1975; 28(7):748-755.
21. Anderson JW, Herman RH. Effect of fasting, caloric restriction, and refeeding on glucose tolerance of normal men. Am J Clin Nutr 1972; 25(1):41-52.
22. Speakman JR. Thrifty genes for obesity and the metabolic syndrome--time to call off the search? Diab Vasc Dis Res 2006; 3(1):7-11.
23. Eaton SB, Strassman BI, Nesse RM et al. Evolutionary health promotion. Prev Med 2002; 34(2):109-118.
24. Neel JV. Diabetes mellitus: a «thrifty» genotype rendered detrimental by «progress»? Am J Hum Genet 1962; 14:353-62.:353-362.
25. Halberg N, Henriksen M, Soderhamn N et al. Effect of intermittent fasting and refeeding on insulin action in healthy men. J Appl Physiol 2005; 99(6):2128-2136.
26. Stannard SR, Johnson NA. Insulin resistance and elevated triglyceride in muscle: more important for survival than ‘thrifty’ genes? J Physiol 2004; 554(3):595-607.
27. Johnson NA, Stannard SR, Rowlands DS et al. Effect of short-term starvation versus high-fat diet on intramyocellular triglyceride accumulation and insulin resistance in physically fit men. Exp Physiol 2006; 91(4):693-703.
28. Adams JM, Pratipanawatr T, Berria R et al. Ceramide content is increased in skeletal muscle from obese insulin-resistant humans. Diabetes 2004; 53(1):25-31.
29. Straczkowski M, Kowalska I, Baranowski M et al. Increased skeletal muscle ceramide level in men at risk of developing type 2 diabetes. Diabetologia 2007; 50(11):2366-2373.
30. Charron MJ, Kahn BB. Divergent molecular mechanisms for insulin-resistant glucose transport in muscle and adipose cells in vivo. J Biol Chem 1990; 265(14):7994-8000.
31. Boden G, Chen X, Ruiz J, White JV, Rossetti L. Mechanisms of fatty acid-induced inhibition of glucose uptake. J Clin Invest 1994; 93(6):2438-2446.
32. Frias JP, Macaraeg GB, Ofrecio J, Yu JG, Olefsky JM, Kruszynska YT. Decreased susceptibility to fatty acid-induced peripheral tissue insulin resistance in women. Diabetes 2001; 50(6):1344-1350.
33. Perseghin G, Scifo P, Pagliato E et al. Gender factors affect fatty acids-induced insulin resistance in nonobese humans: effects of oral steroidal contraception. J Clin Endocrinol Metab 2001; 86(7):3188-3196.
34. Holland WL, Knotts TA, Chavez JA, Wang LP, Hoehn KL, Summers SA. Lipid Mediators of Insulin Resistance. Nutrition Reviews 2007; 65:39-46.
proefschrift Soeters.indb 152 4-8-2008 9:13:28
The Metabolic Response to Fasting in H
umans: a Perspective.
153
Chapter 9
35. Straczkowski M, Kowalska I, Nikolajuk A et al. Relationship between insulin sensitivity and sphingomyelin signaling pathway in human skeletal muscle. Diabetes 2004; 53(5):1215-1221.
36. Serlie MJ, Allick G, Groener JE et al. Chronic Treatment With Pioglitazone Does Not Protect Obese Patients With Diabetes Mellitus Type II from Free Fatty Acid-Induced Insulin Resistance. J Clin Endocrinol Metab 2006; .
37. Serlie MJ, Meijer AJ, Groener JE et al. Short-Term Manipulation of Plasma Free Fatty Acids Does Not Change Skeletal Muscle Concentrations of Ceramide and Glucosylceramide in Lean and Overweight Subjects. J Clin Endocrinol Metab 2007; 92(4):1524-1529.
38. Summers SA. Ceramides in insulin resistance and lipotoxicity. Prog Lipid Res 2006; 45(1):42-72.
39. Powell DJ, Turban S, Gray A, Hajduch E, Hundal HS. Intracellular ceramide synthesis and protein kinase Czeta activation play an essential role in palmitate-induced insulin resistance in rat L6 skeletal muscle cells. Biochem J 2004; 382(Pt 2):619-629.
40. Schmitz-Peiffer C, Craig DL, Biden TJ. Ceramide generation is sufficient to account for the inhibition of the insulin-stimulated PKB pathway in C2C12 skeletal muscle cells pretreated with palmitate. J Biol Chem 1999; 274(34):24202-24210.
41. Turinsky J, Bayly BP, O‘Sullivan DM. 1,2-Diacylglycerol and ceramide levels in rat liver and skeletal muscle in vivo. Am J Physiol 1991; 261(5 Pt 1):E620-E627.
42. Soeters MR, Sauerwein HP, Groener JE et al. Gender-Related Differences in the Metabolic Response to Fasting. J Clin Endocrinol Metab 2007; 92(9):3646-3652.
43. Shadid S, Koutsari C, Jensen MD. Direct Free Fatty Acid Uptake Into Human Adipocytes In Vivo: Relation to Body Fat Distribution. Diabetes 2007; 56(5):1369-1375.
44. Westerbacka J, Corner A, Tiikkainen M et al. Women and men have similar amounts of liver and intra-abdominal fat, despite more subcutaneous fat in women: implications for sex differences in markers of cardiovascular risk. Diabetologia 2004; 47(8):1360-1369.
45. Haymond MW, Karl IE, Clarke WL, Pagliara AS, Santiago JV. Differences in circulating gluconeogenic substrates during short-term fasting in men, women, and children. Metabolism 1982; 31(1):33-42.
46. Magkos F, Patterson BW, Mohammed BS, Klein S, Mittendorfer B. Women Produce Fewer but Triglyceride-Richer Very Low-Density Lipoproteins than Men. J Clin Endocrinol Metab 2007; 92(4):1311-1318.
47. Mittendorfer B. Insulin resistance: sex matters. Curr Opin Clin Nutr Metab Care 2005; 8(4):367-372.
48. Mittendorfer B, Horowitz JF, Klein S. Effect of gender on lipid kinetics during endurance exercise of moderate intensity in untrained subjects. Am J Physiol Endocrinol Metab 2002; 283(1):E58-E65.
49. Blaak E. Gender differences in fat metabolism. Curr Opin Clin Nutr Metab Care 2001; 4(6):499-502.
50. Summers SA, Garza LA, Zhou H, Birnbaum MJ. Regulation of insulin-stimulated glucose transporter GLUT4 translocation and Akt kinase activity by ceramide. Mol Cell Biol 1998; 18(9):5457-5464.
51. Dong LQ, Liu F. PDK2: the missing piece in the receptor tyrosine kinase signaling pathway puzzle. Am J Physiol Endocrinol Metab 2005; 289(2):E187-E196.
52. Scheid MP, Woodgett JR. Unravelling the activation mechanisms of protein kinase B/Akt. FEBS Letters 2003; 546(1):108-112.
53. Stratford S, Hoehn KL, Liu F, Summers SA. Regulation of Insulin Action by Ceramide: Dual mechanisms linking ceramide accumulation to the inhibition of AKT/protein kinase B. J Biol Chem 2004; 279(35):36608-36615.
54. Feng J, Park J, Cron P, Hess D, Hemmings BA. Identification of a PKB/Akt Hydrophobic Motif Ser-473 Kinase as DNA-dependent Protein Kinase. J Biol Chem 2004; 279(39):41189-41196.
55. Soeters MR, Sauerwein HP, Dubbelhuis PF et al. Muscle adaptation to short-term fasting in healthy lean humans. J Clin Endocrinol Metab 2008;jc.
proefschrift Soeters.indb 153 4-8-2008 9:13:28
154
56. Karlsson HKR, Zierath JR, Kane S, Krook A, Lienhard GE, Wallberg-Henriksson H. Insulin-Stimulated Phosphorylation of the Akt Substrate AS160 Is Impaired in Skeletal Muscle of Type 2 Diabetic Subjects. Diabetes 2005; 54(6):1692-1697.
57. Tsintzas K, Chokkalingam K, Jewell K, Norton L, Macdonald IA, Constantin-Teodosiu D. Elevated Free Fatty Acids Attenuate the Insulin-Induced Suppression of PDK4 Gene Expression in Human Skeletal Muscle: Potential Role of Intramuscular Long-Chain Acyl-Coenzyme A. J Clin Endocrinol Metab 2007; 92(10):3967-3972.
58. Stephens FB, Constantin-Teodosiu D, Greenhaff PL. New insights concerning the role of carnitine in the regulation of fuel metabolism in skeletal muscle. J Physiol 2007; 581(2):431-444.
59. Koves TR, Ussher JR, Noland RC et al. Mitochondrial overload and incomplete fatty acid oxidation contribute to skeletal muscle insulin resistance. Cell Metab 2008; 7(1):45-56.
60. Muoio DM, Koves TR. Lipid-induced metabolic dysfunction in skeletal muscle. Novartis Found Symp 2007; 286:24-38; discussion 38-46, 162-3,;%196-203.:24-38.
61. An J, Muoio DM, Shiota M et al. Hepatic expression of malonyl-CoA decarboxylase reverses muscle, liver and whole-animal insulin resistance. Nat Med 2004; 10(3):268-274.
62. Hoppel CL, Genuth SM. Carnitine metabolism in normal-weight and obese human subjects during fasting. Am J Physiol 1980; 238(5):E409-E415.
63. Frohlich J, Seccombe DW, Hahn P, Dodek P, Hynie I. Effect of fasting on free and esterified carnitine levels in human serum and urine: Correlation with serum levels of free fatty acids and [beta]-hydroxybutyrate. Metabolism 1978; 27(5):555-561.
64. Costa CCG, De Almeida IT, Jakobbs Corn, Poll-The B-T, Duran M. Dynamic Changes of Plasma Acylcarnitine Levels Induced by Fasting and Sunflower Oil Challenge Test in Children. [Miscellaneous Article]. Pediatric Research 1999; 46(4):440.
65. Spriet LL, Tunstall RJ, Watt MJ, Mehan KA, Hargreaves M, Cameron-Smith D. Pyruvate dehydrogenase activation and kinase expression in human skeletal muscle during fasting. J Appl Physiol 2004; 96(6):2082-2087.
66. Osler ME, Zierath JR. Minireview: Adenosine 5’-Monophosphate-Activated Protein Kinase Regulation of Fatty Acid Oxidation in Skeletal Muscle. Endocrinology 2008; 149(3):935-941.
67. Randle PJ, Garland PB, Hales CN, Newsholme EA. The glucose fatty-acic cycle: its role in insulin sensitivity and the metabolic disturbances of diabetes mellitus. The Lancet 1963; 281(7285):785-789.
68. Boden G, Shulman GI. Free fatty acids in obesity and type 2 diabetes: defining their role in the development of insulin resistance and beta-cell dysfunction. Eur J Clin Invest 2002; 32 Suppl 3:14-23.
69. Webber J, Taylor J, Greathead H, Dawson J, Buttery PJ, Macdonald IA. Effects of fasting on fatty acid kinetics and on the cardiovascular, thermogenic and metabolic responses to the glucose clamp. Clin Sci (Lond) 1994; 87(6):697-706.
70. Fukao T, Lopaschuk GD, Mitchell GA. Pathways and control of ketone body metabolism: on the fringe of lipid biochemistry. Prostaglandins, Leukotrienes and Essential Fatty Acids 2004; 70(3):243-251.
71. Laffel L. Ketone bodies: a review of physiology, pathophysiology and application of monitoring to diabetes. Diabetes Metab Res Rev 1999; 15(6):412-426.
72. Halestrap A, Meredith D. The SLC16 gene familyGÇöfrom monocarboxylate transporters (MCTs) to aromatic amino acid transporters and beyond. Pflügers Archiv European Journal of Physiology 2004; 447(5):619-628.
73. Hack A, Busch V, Pascher B et al. Monitoring of ketogenic diet for carnitine metabolites by subcutaneous microdialysis. Pediatr Res 2006; 60(1):93-96.
74. Swiatek KR, Dombrowski GJ, Chao KL, Chao HL. Metabolism of - and -3-hydroxybutyrate by rat liver during development. Biochemical Medicine 1981; 25(2):160-167.
75. Hildebrandt LA, Spennetta T, Elson C, Shrago E. Utilization and preferred metabolic pathway of ketone bodies for lipid synthesis by isolated rat hepatoma cells. Am J Physiol Cell Physiol 1995; 269(1):C22-C27.
proefschrift Soeters.indb 154 4-8-2008 9:13:28
The Metabolic Response to Fasting in H
umans: a Perspective.
155
Chapter 9
76. Hildebrandt L, Spennetta T, Ackerman R, Elson C, Shrago E. Acyl CoA and Lipid Synthesis from Ketone Bodies by the Extramitochondrial Fraction of Hepatoma Tissue. Biochemical and Biophysical Research Communications 1996; 225(1):307-312.
77. Molven A, Matre GE, Duran M et al. Familial Hyperinsulinemic Hypoglycemia Caused by a Defect in the SCHAD Enzyme of Mitochondrial Fatty Acid Oxidation. Diabetes 2004; 53(1):221-227.
78. Soeters MR, Sauerwein HP, Duran M. Muscle D-3-hydroxybutyrylcarnitine: an alternative pathway in ketone body metabolism. Unpublished Work, chapter 5 of this thesis
79. Ramsay RR, Gandour RD, van der Leij FR. Molecular enzymology of carnitine transfer and transport. Biochim Biophys Acta 2001; 1546(1):21-43.
80. Lopaschuk GD, Belke DD, Gamble J, Toshiyuki I, Schonekess BO. Regulation of fatty acid oxidation in the mammalian heart in health and disease. Biochimica et Biophysica Acta (BBA) - Lipids and Lipid Metabolism 1994; 1213(3):263-276.
81. Stannard SR, Thompson MW, Fairbairn K, Huard B, Sachinwalla T, Thompson CH. Fasting for 72 h increases intramyocellular lipid content in nondiabetic, physically fit men. Am J Physiol Endocrinol Metab 2002; 283(6):E1185-E1191.
82. Jensen MD, Ekberg K, Landau BR. Lipid metabolism during fasting. Am J Physiol Endocrinol Metab 2001; 281(4):E789-E793.
83. Varady KA, Hellerstein MK. Alternate-day fasting and chronic disease prevention: a review of human and animal trials. Am J Clin Nutr 2007; 86(1):7-13.
84. Heilbronn LK, Civitarese AE, Bogacka I, Smith SR, Hulver M, Ravussin E. Glucose tolerance and skeletal muscle gene expression in response to alternate day fasting. Obes Res 2005; 13(3):574-581.
85. Heilbronn LK, Smith SR, Martin CK, Anton SD, Ravussin E. Alternate-day fasting in nonobese subjects: effects on body weight, body composition, and energy metabolism. Am J Clin Nutr 2005; 81(1):69-73.
86. Fryburg DA, Barrett EJ, Louard RJ, Gelfand RA. Effect of starvation on human muscle protein metabolism and its response to insulin. Am J Physiol Endocrinol Metab 1990; 259(4):E477-E482.
87. Tsalikian E, Howard C, Gerich JE, Haymond MW. Increased leucine flux in short-term fasted human subjects: evidence for increased proteolysis. Am J Physiol Endocrinol Metab 1984; 247(3):E323-E327.
88. Fery F, Plat L, Melot C, Balasse EO. Role of fat-derived substrates in the regulation of gluconeogenesis during fasting. Am J Physiol Endocrinol Metab 1996; 270(5):E822-E830.
89. Dann SG, Thomas G. The amino acid sensitive TOR pathway from yeast to mammals. FEBS Letters 2006; 580(12):2821-2829.
90. Moller L, Jorgensen HS, Jensen FT, Jorgensen JO. Fasting in healthy subjects is associated with intrahepatic accumulation of lipids as assessed by 1H-magnetic resonance spectroscopy. Clin Sci (Lond) 2007.
91. Knapik JJ, Meredith CN, Jones BH, Suek L, Young VR, Evans WJ. Influence of fasting on carbohydrate and fat metabolism during rest and exercise in men. J Appl Physiol 1988; 64(5):1923-1929.
92. Landau BR, Wahren J, Chandramouli V, Schumann WC, Ekberg K, Kalhan SC. Contributions of gluconeogenesis to glucose production in the fasted state. J Clin Invest 1996; 98(2):378-385.
93. Krssak M, Falk PK, Dresner A et al. Intramyocellular lipid concentrations are correlated with insulin sensitivity in humans: a 1H NMR spectroscopy study. Diabetologia 1999; 42(1):113-116.
94. Hojlund K, Wildner-Christensen M, Eshoj O et al. Reference intervals for glucose, beta-cell polypeptides, and counterregulatory factors during prolonged fasting. Am J Physiol Endocrinol Metab 2001; 280(1):E50-E58.
95. Wood IS, Trayhurn P. Glucose transporters (GLUT and SGLT): expanded families of sugar transport proteins. Br J Nutr 2003; 89(1):3-9.
proefschrift Soeters.indb 155 4-8-2008 9:13:28
proefschrift Soeters.indb 156 4-8-2008 9:13:29
English SummaryNederlandse SamenvattingDankwoordBiografie10
proefschrift Soeters.indb 157 4-8-2008 9:13:29
proefschrift Soeters.indb 158 4-8-2008 9:13:29
English Summ
ary
159
Chapter 10
English Summary
In this thesis, we investigated the metabolic adaptation to fasting. Investigating pathways,
involved in protecting the body from energy depletion, may provide us with answers on
the metabolic adaptation to overfeeding, i.e. the metabolic consequences of obesity.
The background of this thesis is presented in chapter 1. General considerations (e.g.
definitions) on fasting are discussed. Consequently, the adaptations to fasting of fatty
acid and glucose metabolism are reviewed. In the section on the adaptation of glucose
metabolism, much attention is paid to peripheral insulin stimulated glucose uptake and
the role of fatty acids in this process. Finally, the thesis outline is presented.
In chapter 2 we studied whether the relative protection from FFA-induced insulin
resistance during fasting in women, is associated with lower muscle ceramide
concentrations compared to men, since women have lower plasma glucose levels than
men despite higher plasma FFA during fasting, suggesting protection from FFA-induced
insulin resistance.
We studied lean men and women in hyperinsulinemic euglycemic clamp studies. After
a 38 h fast, plasma glucose levels were significantly lower in women than men with a
trend for a lower endogenous glucose production in women, while FFA and lipolysis
were significantly higher. Insulin-mediated peripheral glucose uptake was not different
between sexes. There was no gender difference in muscle ceramide in the basal state
and ceramide did not correlate with peripheral glucose uptake. Muscle FAT/CD36 was
not different between sexes in the basal state and during the clamp.
These data show that during fasting, women are relatively protected from FFA-induced
insulin resistance possibly by preventing myocellular accumulation of ceramide. This is
not explained by differences in total muscle FAT/CD36.
It has been demonstrated repeatedly that short-term fasting induces insulin resistance
although the exact mechanism in humans is unknown to date. Muscle ceramide is
suggested to induce insulin resistance by interfering with the insulin signaling cascade
in obesity. In chapter 3 we performed clamp studies to investigate peripheral insulin
sensitivity together with muscle ceramide concentrations and protein kinase B/AKT
phosphorylation after short-term fasting.
We found that insulin mediated peripheral glucose uptake was significantly lower
after 62 h compared to 14 h of fasting. Intramuscular ceramide concentrations tended to
proefschrift Soeters.indb 159 4-8-2008 9:13:29
160
increase during fasting and that the phosphorylation of protein kinase B/AKT at serine473
in proportion to the total amount of protein kinase B/AKT was significant lower during
the clamp. Muscle ceramide did not correlate with plasma free fatty acids.
This demonstrates that fasting decreases insulin mediated peripheral glucose uptake
with lower phosphorylation of AKT at serine473, which may play a regulatory role in
fasting induced insulin resistance. The role of muscle ceramide remains elusive.
The transition from the fed to the fasted resting state is characterized by changes in lipid
metabolism besides peripheral insulin resistance. Acylcarnitines have been suggested
to play a role in insulin resistance besides other long-chain fatty acid metabolites. It is
unknown whether muscle long-chain acylcarnitines increase during fasting or relate with
glucose/fat oxidation and insulin sensitivity in lean healthy humans.
Chapter 4 discusses hyperinsulinemic euglycemic clamp studies after 14 and 62 hours
of fasting. Hyperinsulinemia decreased long-chain muscle acylcarnitines after 14 but not
after 62 hours of fasting. During the basal state and clamp, fatty acid oxidation was
lower after 14 vs. 62 hours of fasting.Absolute changes in glucose and fat oxidation
within clamps were not different. Muscle long-chain acylcarnitines did not correlate with
substrate oxidation or insulin-mediated peripheral glucose uptake.
It was concluded that muscle long-chain acylcarnitines do not unconditionally reflect
fatty acid oxidation. The adaptation in fatty acid oxidation suggests a different insulin-
regulated set point.
The ketone body D-3-hydroxybutyrate (D-3HB) plays an important role in the adaptation
to fasting. It has been suggested by at least two separate studies that D-3HB can be
coupled to carnitine to form 3-hydroxybutyrylcarnitine. However, it is unknown whether
and how D-3HB can be coupled to carnitine resulting in D-3HB-carnitine in humans.
To assess which stereo isomers of 3-hydroxybutyrylcarnitine are present in vivo, we
performed pancreatic clamp studies in 12 healthy men and homogenate studies in liver
and muscle of mice as described in chapter 5.
Muscle D-3HB-carnitine was approximately 7.5 fold higher compared to L-3HB-
carnitine in ketotic men after 38 h of fasting. Muscle D-3HB-carnitine and D-3-HB turnover
correlated significantly. The ACS pathway was active in liver and muscle homogenates
though less than the SCOT pathway that was only active in muscle homogenates.
Therefore, D-3HB-carnitine, related with the D-3HB turnover, can be formed in
muscle by mitochondrial SCOT and a carnitine acyltransferase. These data provide a
proefschrift Soeters.indb 160 4-8-2008 9:13:29
English Summ
ary
161
Chapter 10
newly identified alternative pathway of ketone body metabolism. The purpose of D-3HB-
carnitine synthesis and its fate remain to be elucidated.
The dose-response relationship between insulin and ketone bodies was demonstrated
to be shifted to the right in type 2 diabetes mellitus patients. In contrast ketone body
levels have also been reported to be decreased in obesity. To clarify this paradox, we
investigated the metabolic adaptation to fasting with respect to glucose and ketone
body metabolism in lean and obese men without non insulin dependent diabetes mellitus
in chapter 6. We hypothesized ketone body production to be equal under equal plasma
insulin levels thereby reflecting absence of insulin resistance on ketogenesis.
Pancreatic clamp studies were performed after 38 hours of fasting in lean and obese
men. In the basal state, ketone body fluxes were higher in lean compared with obese
men. During the pancreatic clamp, no differences were found in ketone body fluxes
during similar plasma insulin levels. Peripheral glucose uptake was lower in obese men.
Obese subjectsare resistant to insulin’s effect on stimulation glucose of glucose
uptake, but not its inhibiting effect on ketogenesis, implying differential insulin sensitivity
of intermediary metabolism in obesity.
Intermittent fasting has shown to increase insulin sensitivity but it is uncertain whether
intermittent fasting selectively influences intermediary metabolism. Such selectivity might
be advantageous when adapting to periods of food abundance and food shortage.
In chapter 7 we investigated the effects of 2 weeks intermittent fasting vs. 2 weeks
of a standard diet on glucose, lipid and protein metabolism in the basal state and during
a two-step hyperinsulinemic euglycemic clamp with assessment of energy expenditure
and muscle insulin signaling.
Peripheral glucose uptake during step 1, but not step 2, was significantly higher
after intermittent fasting compared to SD but this was not the case for hepatic insulin
sensitivity. Lipolysis was more inhibited by insulin after intermittent fasting. Proteolysis
was only lower after intermittent fasting during step 2. Intermittent fasting decreased
resting energy expenditure and tended to augment insulin signaling of AKT, whereas
higher phosphorylation of glycogen synthase was found.
Intermittent fasting differentially affects glucose, lipid and protein metabolism. The
decrease in REE after intermittent fasting is a potential cause for increasing weight during
intermittent fasting when caloric intake is not adjusted. Whether intermittent fasting is
beneficial in improving peripheral insulin resistance in obese insulin resistant subjects
remains to be established.
proefschrift Soeters.indb 161 4-8-2008 9:13:29
162
The interpretation of the prolonged supervised fast (to detect an insulinoma) is troublesome
in those patients who develop hypoglycemia with appropriate hypoinsulinemia during
this test. In chapter 8 we investigated in this group of patients whether abnormalities
in intermediary metabolism, known for their relation with ketotic hypoglycemia (fatty
acid oxidation and amino/organic acids) could be detected which might explain the
hypoinsulinemic hypoglycemia.
We studied 10 patients with otherwise unexplained hypoglycemia during prolonged
fasting in an extended metabolic diagnostic protocol based on stable isotope techniques
after an overnight fast.
There were no hypoglycemic events. No abnormalities in fatty acid oxidation (FAO) or
in amino acid/organic acids were found in this patient group.
We found no signs of metabolic derangements in these patients. Therefore, the
previously observed low plasma glucose values during the supervised fast probably
represent the lower tail of the Gaussian Curve of plasma glucose concentrations during
fasting.
Finally, chapter 9 is a perspective describing the integrated metabolic response to
fasting regarding adaptive changes in lipid and glucose metabolism. Here we tried to
elucidate the relevance and physiological aspects of a mechanism needed for survival of
the organism.
Although the purpose of fasting may be simple (i.e. to safe energy), the exact reason
for these adaptations is unknown. We shed light on the discrepancy between muscle
FAO and lipid supply during short-term fasting.Moreover we discuss fasting induced
peripheral insulin resistance. Although this is a highly regulated process as shown in this
thesis, its explanation is currently lacking. The thought that this insulin resistance simply
prevents hypoglycemia may be too feeble. Some suggestions are put forward for future
studies that may increase our understanding of the physiological adaptation to fasting as
well as pathophysiological states as obesity induced insulin resistance and type 2 diabetes
mellitus.
proefschrift Soeters.indb 162 4-8-2008 9:13:29
163
Chapter 10
Nederlandse Sam
envatting
Nederlandse Samenvatting
In dit proefschrift hebben wij de Metabole aanpassing aan vasten onderzocht. Het
onderzoeken van paden die betrokken zijn bij het voorkomen van energie verlies, kan
leiden tot het begrijpen van de metabole aanpassing aan overvoeden ofwel de metabole
gevolgen van obesitas. De achtergrond van dit proefschrift is besproken in hoofdstuk
1. Algemene overwegingen (o.a. definities) passeren de revue. Vervolgens komen de
aanpassingen in vet en glucose metabolisme tijdens vasten aan de orde. In de paragraaf
over glucose stofwisseling wordt nadruk gelegd op de perifere insuline gestimuleerde
glucose opname én de rol van vrije vetzuren daarin. Als laatste wordt de opbouw van
het proefschrift gepresenteerd.
In hoofdstuk 2 hebben wij onderzocht of de relatieve bescherming van vrouwen tegen
het insuline resistente vermogen van vrije vetzuren geassocieerd is met lagere ceramide
spiegels in skeletspier. Vrouwen hebben namelijk lagere plasma glucose waarden
vergeleken met mannen ondanks hogere plasma vrije vetzuur spiegels. Dit suggereert
bescherming tegen vrije vetzuur geïnduceerde insuline resistentie.
Wij bestudeerden slanke mannen en vrouwen tijdens een hyperinsulinemische
euglycemische clamp. In de basale staat hadden vrouwen na 38 uur vasten significant
lagere glucose spiegels met een trend tot lagere endogene glucose productie. Plasma
vrije vetzuren en lipolyse waren daarentegen significant hoger bij vrouwen. Insuline
gestimuleerde glucose opname verschilde niet tussen de beide geslachten. Er was geen
geslachtsverschil wat betreft ceramide in spier in de basale staat, noch correleerde
ceramide met vrije vetzuren of perifere glucose opname. Spier concentraties van de
vetzuurtransporter FAT/CD36 waren niet verschillend in de basale staat of clamp.
Deze data laten zien dat tijdens vasten vrouwen relatief beschermd zijn tegen de
insuline resistente effecten van vrije vetzuren, mogelijk doordat accumulatie van ceramide
wordt voorkomen. Dit is niet verklaard door verschillen in spier FAT/CD36.
Kortdurend vasten induceert insuline resistentie, hoewel het pathofysiologische
mechanisme nog onduidelijk is. Er wordt gedacht dat ceramide in spier verantwoordelijk is
voor het induceren van insuline resistentie via remming van de insuline signaal transductie
in obesitas. In hoofdstuk 3 hebben wij clamp studies beschreven waarin werd gekeken
naar perifere insuline gevoeligheid, ceramide en proteïne kinase B/AKT fosforylering in
spier na kortdurend vasten
proefschrift Soeters.indb 163 4-8-2008 9:13:29
164
Insuline gemedieerde glucose opname was significant lager na 62 uur vasten
vergeleken met 14 uur vasten. Spier ceramide concentraties lieten een trend tot hogere
waarden zien na vasten en de fosforylering van AKT (serine473) in proportie tot totaal
AKT was significant lager na vasten tijdens de clamp. Ceramide in spier correleerde niet
met de plasma vrije vetzuren.
Vasten induceert insuline resistentie met betrekking tot glucose opname met een
verlaagde fosforylering, en dus regulerende rol, van AKT. De definitieve rol van ceramide
blijft vooralsnog onbekend
De overgang van de gevoede naar de gevaste staat wordt gekarakteriseerd door
veranderingen in vetstofwisseling (naast het ontstaan van insuline resistentie). Het
is verondersteld dat acylcarnitinen een rol spelen in insuline resistentie naast andere
metabolieten van lange keten vetzuren. Het is onbekend of de concentratie lange keten
acylcarnitinen in spier tijdens vasten hoger wordt en of er een relatie is met glucose/
vetoxidatie en insuline gevoeligheid in slanke gezonde vrijwilligers.
Hoofdstuk 4 bespreekt de hyperinsulinemische euglycemische clamp studies na 14
en 62 uur vasten. Hyperinsulinemie verlaagt lange keten acylcarnitinen na 14 uur vasten
maar niet na 62 uur vasten. Tijdens de basale staat en de clamp,was de vet oxidatie
lager na 14 uur vasten maar de absolute verschillen in glucose en vet oxidatie waren
niet verschillend. Spier acylcarnitinen correleerden niet met substraat oxidatie of insuline
gemedieerde glucose opname.
Lange keten acylcarnitinen in spier reflecteren niet de vetzuuroxidatie en de aanpassing
in vetzuuroxidatie na vasten maakt een ander set point aannemelijk.
Het ketonlichaam D-3-hydroxybutyraat (D-3HB) speelt een belangrijke rol in de adaptatie
aan vasten. Twee eerdere studies suggereerden dat D-3HB kan koppelen aan carnitine om
3-hydroxybutyrylcarnitine te vormen. Het is echter onbekend of en hoe D-3HB-carnitine
synthese verloopt in de mens.
Om te bepalen welke stereo isomeren van 3-hydroxybutyrylcarnitine aanwezig zijn in
vivo, verichtten wij pancreatische clamps bij 12 gezonde mannen en in vitro studies in
lever en spier van muizen (hoofdstuk 5).
Spier D-3HB-carnitine was ongeveer 7.5 maal hoger vergeleken met L-3HB-carnitine
in ketotische mannen na 38 uur vasten. Spier D-3HB-carnitine and D-3-HB turnover
correleerden significant. Het acetyl-CoA synthase was actief in lever en spier homogenaten
maar minder dan het succinyl-CoA-oxo-transferase.(SCOT) pad dat enkel actief was in
spier homogenaten.
proefschrift Soeters.indb 164 4-8-2008 9:13:29
165
Chapter 10
Nederlandse Sam
envatting
Wij hebben geconcludeerd dat D-3HB-carnitine is gerelateerd aan D-3HB turnover
en kan worden gesynthetiseerd in spier door mitochondriaal SCOT en een carnitine
acyltransferase: een alternatief pad in ketonlichaam metabolisme dat tot nog toe
onbekend was. Het reden voor D-3HB-carnitine synthese alsmede het lot van D-3HB-
carnitine zijn vooralsnog onbekend.
De dosis-respons relatie tussen insuline and ketonlichamen zou naar rechts zijn verschoven
in niet insuline afhankelijke diabetes mellitus (NIDDM). Echter,
plasma ketonlichaam spiegels zijn lager in obesitas. In hoofdstuk 7 onderzochten wij de
metabole respons op vasten met betrekking tot glucose en ketonlichaam metabolisme
in slanke en obese mannen zonder NIDDM. We verwachtten dat ketonlichaam productie
niet verschillend was bij gelijke plasma insuline spiegels. Dit zou de afwezigheid van
insuline resistentie van ketogenese betekenen.
Pancreatische clamps werden uitgevoerd na 38 uur vasten in slanke en obese
mannen. In de basale staat, was de ketonlichaamproductie hoger in de slanke mannen
vergeleken met de obese mannen. Tijdens de clamp waren er geen verschillen in
ketonlichaamproductie onder gelijke insuline spiegels. Perifere glucose opname was
echter lager in de obese mannen.
Deze data wijzen op differentiële insuline gevoeligheid van het intermediaire
metabolisme in obesitas.
Intermitterend vasten verhoogt de perifere insuline gevoeligheid. Het is echter onbekend
of intermitterend vasten selectief het intermediaire metabolisme beïnvloedt. Zulke
selectiviteit kan voordelig zijn in de adaptatie aan verhoogde en verlaagde voedsel
inname.
In hoofdstuk 7 onderzochten wij de effecten van twee weken intermitterend vasten
(IF) vergeleken met twee weken standaard isocalorisch dieet (SD) op glucose, vet en
eiwit metabolisme (in de basale staat en tijdens een twee-staps hyperinsulinemische
euglycemische clamp) en energieverbruik.
Perifere glucose opname was hoger tijdens stap 1 maar niet tijdens stap 2 na IF
vergeleken met het SD, maar dit was niet geval voor hepatische insuline sensitiviteit.
Lipolyse was sterker onderdrukt door insuline na IF. Proteolyse was lager tijdens stap
2 na IF. Energieverbruik was lager na IF. Er was een trend tot lagere fosforylering van
AKT (serine473) met een significant hogere fosforylering van glycogeen synthase kinase
(serine9).
proefschrift Soeters.indb 165 4-8-2008 9:13:29
166
IF heeft verschillende effecten op glucose, vet en eiwit metabolisme. De afname van
energieverbruik waarschuwt voor gewichtstoename tijdens IF als de calorische inname
niet wordt aangepast. Of IF gunstig is om insuline gevoeligheid te verbeteren is daarom
nog maar zeer de vraag.
Het interpreteren van de verlengde vastentest (ter detectie van een insulinoom) is
moeilijk in patiënten met een hypoglycemie en hypoinsulinemie. In hoofdstuk 8 hebben
wij onderzocht in deze groep patiënten of er afwijkingen bestaan in het intermediaire
metabolisme, zoals vetzuuroxidatie stoornissen en afwijkingen in aminozuren en/
of organische zuren. Zulke afwijkingen kunnen hypoinsulinemische hypoglycemiën
verklaren.
Wij onderzochten 10 patienten met onverklaarde hypoinsulinemische hypoglycemiën
tijdens de verlengde vastentest in een uitgebreid metabool diagnostisch protocol
(na een overnachtse vast) gebaseerd op stabiele isotoop technieken. Er waren geen
hypoglycemiën. Tevens werden er geen defecten in de vetzuuroxidatie of profielen van
aminozuren en organische zuren waargenomen.
Aangezien er geen afwijkingen werden gevonden, geven deze hypoinsulinemische
hypoglycemiën mogelijk het onderste bereik van de Gauss curve weer.
In het perspective, hoofstuk 9, beschrijven wij de geïntegreerde metabole respons op
vasten met betrekking tot veranderingen in vet en glucose metabolisme. Tevens zoeken
wij een verklaring voor de relevantie en fysiologische aspecten van dit mechanisme dat
nodig is voor de overleving van het organisme.
Hoewel het doel van vasten simpel mag zijn (namelijk het sparen van energie), is
de exacte reden van deze aanpassingen onbekend. Wij hebben ons gericht op de
discrepantie tussen vetoxidatie en vet toevoer tijdens vasten enerzijds en de vasten
geïnduceerde insuline resistentie anderzijds. Hoewel de aanpassing aan vasten sterk
gereguleerd is, blijft de verklaring ontbreken. De gedachte dat insuline resistentie tijdens
vasten simpelweg hypoglycemiën voorkomt lijkt te simpel. Afsluitend doen wij een aantal
suggesties voor toekomstige onderzoeksprojecten die niet alleen kunnen leiden tot een
beter begrip van de fysiologische aanpassing aan vasten maar ook tot een beter begrip
van pathosfysiologische omstandigheden als obesitas geïnduceerde insuline resistentie
en type 2 diabetes mellitus.
proefschrift Soeters.indb 166 4-8-2008 9:13:29
167
Chapter 10
Dankw
oord
DANKWOORD
Soms heerst er twijfel bij een promovendus over het dankwoord omdat er zoveel mensen
betrokken zijn geweest bij het tot stand komen van het proefschrift. Soms is er aarzeling
of het woord ”dankwoord” wel de juiste lading dekt van de waarheid of van het gevoel
van de jonge onderzoeker zelf.
Hoewel ik het schrijven van dit proefschrift nooit ervaren heb als de grootste last of
opgave, weerspiegelen deze laatste regels mijn waardering, bewondering, maar soms ook
diepe liefde voor hen die gaven wat nodig was. Variërend van een achteloze opmerking
over een proef of manuscript tot het uitgebreide redigeren of het onvoorwaardelijke “er
zijn”.
Vrijwilligers maken het onderzoek van de Metabole Groep mogelijk en dit boekje was er
niet zonder Antonius, Christina, Claire, Daniël, Dennis, Diederik, Dutton, Eduard, Emiel,
Erik, Erwin, Floor, Ieke, Jan, Jaques, Jelle, Jeroen, Jesse, Jitske, Jochem, Joost, Joost,
Joost, Jorrit, Karel, Kimo, Leon, Loes, Maarten, Maren, Marlinde, Martijn, Matthijs,
Micha, Michiel, Miro, Nico, Nina, Onno, Philip, Pim, Ralph, Renato, Robert, Robert,
Robert, Roland, Ruben, Sander, Sandy, Sarah, Serge, Sietse, Sigrid, Theo, Tiemen, Tim,
Vincent, Waldo, Wessel en Wouter. Tevens gaat mijn dank uit naar de hypoglycemie
patiënten die ik soms moest “teleurstellen” omdat wij geen oorzaak vonden voor de lage
bloedsuiker waarden of klachten die zij hadden.
Hooggeleerde Prof. dr. H.P. Sauerwein, waarde Hans; mijn eerste herinnering aan jou is
een kleine doch statige man, gekleed in zwarte pantalon, dito hemd met fluorescerende
kleurtjes en roze Doctor Martens©. Ik realiseer mij steeds meer dat wij niet altijd hetzelfde
zijn of denken, maar dat heeft naar mijn smaak ook de chemie gebracht. Ik ben trots
dat ik mij gedurende twee jaar onder jouw hoede fulltime aan research heb mogen
wijden. Deze waarlijk luxe situatie heeft niet enkel geleid tot mooie projecten, maar ook
tot reizen naar andere continenten. Jij toont dat er geen positieve correlatie is tussen
lichaamslengte en Helicopter View1. Tevens hoop ik op een dag het door jou verfoeide
monosynaptisme2 achter mij te laten. Wellicht dat transatlantische sturing uit St. Louis
mogelijk is!
Weledelzeergeleerde Dr. M.J. Serlie, waarde Mireille; eigenlijk was jij het die in eerste
instantie vroeg of ik de Metabole Groep wilde versterken. Een volgende keer zou ik het
1 Het vermogen een detail in de grotere context te plaatsen (Sauerweiniaans).2 De beperking maar één conclusie te destilleren uit een waarneming (Sauerweiniaans).
proefschrift Soeters.indb 167 4-8-2008 9:13:29
168
weer doen en daar speelt jouw visie zeker een rol in. Natuurlijk heb jij, door Hans, ook
Helicopter View, maar méér nog weet je als geen ander tot de kern van de zaak door te
dringen, zowel in je redigeren van manuscripten als tijdens het superviseren van de 314.
De taak die je jezelf voor de komende jaren hebt opgelegd is niet de minste en ik heb
een rotsvast vertrouwen in je.
Weledelzeergeleerde Dr. ir. M.T. Ackermans, waarde Mariëtte; zo belangrijk als de
isotopen zijn voor onze studies, zo belangrijk ben jij voor de Metabole Groep. Vanaf
het begin heb ik mij altijd welkom gevoeld op F2 hoewel ik nog nooit één verrijking
heb gemeten. En als dat er echt niet meer van komt… dan gaan we toch gewoon
schaatsen?
Graag wil ik Prof. dr. E. Blaak, Prof. dr. F. Kuipers, Prof. dr. J.A. Romijn, Prof. dr. M.M. Levi,
Prof. dr. W.M. Wiersinga en Prof. dr. F.A. Wijburg danken voor het kritisch bestuderen
van het manuscript en het zitting nemen in de promotiecommissie.
Er is altijd een spanningsveld tussen opleiding en onderzoek, maar het bijzondere
opleidingsklimaat in het AMC, gecreëerd door Prof. dr. M. Levi, Prof. dr. P. Speelman en
Prof. dr. J.B. Hoekstra geeft veel ruimte tot zelfontplooiing. Dit geldt voor zowel research
als het intussen beruchte skiweekeinde.
Hooggeleerde Prof. dr. E. Fliers, beste Eric; dit proefschrift is deels geschreven tijdens de
eerste maanden van mijn aandachtsgebied Endocrinologie en Metabolisme. Ik waardeer
het zeer dat dit mogelijk was en ik denk dat dit mij ook gestimuleerd heeft om “door
te pakken”. Op een koude zeilmiddag op het IJsselmeer werd het idee voor het D2-stuk
geboren. Nu deze joint-venture met Leiden en Rotterdam bijna publicatierijp is, volgt een
ander endocrien staartje van dit proefschrift!
Ik ben trots op de samenwerking met de diverse afdelingen binnen en buiten het AMC.
Onze belangrijkste partners zijn natuurlijk het STabiele Isotopen Lab en Endocrinologie
Lab. Beste An, ik heb de snoeptrommel al veel vaker leeggegeten dan aangevuld. En
eerlijk is eerlijk: ik hoop dat in de toekomst ook zo te blijven doen! Beste Barbara,
ondanks jouw vertrek hoor jij in dit dankwoord. Ik dank jullie beiden voor het meten van
al die verrijkingen; mijn waardering is enorm. Erik, hoewel wij niet heel veel rechtstreeks
contact hadden over de studies, weet ik dat je op de achtergrond zeer betrokken was!
De Medische Biochemie (ceramide en insuline signaleringspaden): Geachte Prof.
dr. J.M. Aerts, Dr. J.E. Groener en Dr. P.F. Dubbelhuis, beste Hans, Ans en Peter, jullie
proefschrift Soeters.indb 168 4-8-2008 9:13:29
169
Chapter 10
Dankw
oord
constructieve manier van werken heeft mij altijd goed gedaan. Waarde Aldi, de ceramide
waarden komen rechtstreeks van jou! Ik denk aan je.
Het lab Genetisch Metabole Ziekten (o.a. (acyl)carnitine analyses): Geachte Prof. dr.
R.J. Wanders en Dr. M. Duran, beste Ron en Ries, het is lastig om een bespreking met
jullie te beleggen… en als het dan eindelijk lukt, zijn jullie onnavolgbaar in de metabole
paden en stofwisselingsziekten. Jullie enthousiasme blijkt besmettelijk. Beste Sander,
geachte Dr. S.M. Houten, wij zaten samen in de brugklas. De manier waarop jij nuanceert
en zelfkritisch bent, moet een voorbeeld zijn voor anderen. Tevens is het mooi om te zien
hoe jij langzaam maar zeker jouw zachte G verliest… Jos, de homogenaat studies zijn de
kers op de hydroxybutyrylcarnitine-taart!
Lieve Cora, het intermitterend vasten project was niet zo uit de verf gekomen zonder
jouw enthousiasme. Je bent doortastend en altijd al weer een paar stappen verder met
je gedachten. Ik hoop nog veel van je te leren!
Dit proefschrift is grotendeels een AMC productie, maar er waren ook twee
buitengewesten: Geachte Prof. dr. J.F. Glatz, beste Jan, jouw deur (CARIM, Maastricht,
FAT/CD36 analyses) stond altijd open. Het is bijzonder als het contact zo laagdrempelig
en vanzelfsprekend is. Dear Professor C.L. Hoppel (Case Western Reserve University,
Cleveland, Verenigde Staten), I would like to thank you for putting a great effort in
determining L- and D-stereo isomers of hydroxybutyrylcarnitine.
De afdeling Endocrinologie en Metabolisme is een plek waar ik mij erg thuis voel. Marga,
Birgit en Marlies; jullie vorm(d)en de spil van zorg! Ik ga heus nog wel eens keer mee
lunchen, maar dan wel eerst een gratis nagelstyling van Marga! Martine, dank voor al die
rust en uitleg aan de studenten wat betreft OGTT’s, infusen en héél véél ander geregel:
werkelijk top! Waarde Peter, kun je nog een keer de O2-consumtie van de ketogenese
uitleggen? Xander en Nadia, jullie hebben geen metabool onderwerp en daarom wél
veel verstand van SPSS en statistiek. Gabor (honorarium voor nö 11 volgt!) en Eefje, soms
moet de deur van 166 even dicht: wanneer komt die koelkast met bier?
Beste Lars, Saskia, Regje, Gideon, Mirjam en Hidde, wij beschouwden onszelf als de
metabole onderzoekers. De “Diners Métaboliques” waren sporadisch, maar altijd
zeer goed voor hart, ziel en maag... Intussen delen we ervaringen over de opleiding,
hypotheken, kinderen en de liefde.
Saskia en Regje, 11 september 2008 dus… Lieve Saskia, jij hebt, na suggestie van
een van jouw opponenten, het balletje aan het rollen gekregen en het is het gelukt!
Deze Triplo-Promotie is uniek! Lieve Bloem, ik heb veel aan je te danken. In mijn eerste
proefschrift Soeters.indb 169 4-8-2008 9:13:29
170
maanden op F5 heb jij mij enorm geholpen met werkelijk van alles… Gaan we nu eindelijk
eens een goede borrel drinken?
Een jongere generatie onderzoekers staat klaar. Beste Linda en Martijn, studenten
van het eerste uur, dankzij jullie is de KETO-studie een succes geworden. Lieve Nicolette,
jij maakte deel uit van mijn tweede lichting studenten; ik ben trots dat je nu bij Hans en
Mireille gaat promoveren. Myrthe, goede wijn behoeft geen krans, jouw assertiviteit en
werklust dwingen respect af. De Metabole Groep is springlevend!
Waarde Heeren van de Pelikaan, Boris, Mark, Ernst, Joost, Djan, Jochem en Jelger. In 1997
hield onze acht op te bestaan en nog steeds zijn wij onafscheidelijk. Onze onderwerpen
zijn minder slap, Mark stelt geen medisch-erotische vragen meer en ons hoofd wordt niet
meer gewogen: worden we toch volwassen? Jelger, jij en ik in de 2- op de Bosbaan….
Dat gaat nooit meer weg.
Merijn, of je nu wilt of niet, je bent mijn jongste broertje, een Benjamin van deze tijd.
www.merijnsoeters.com is niet zonder toeval een succes: jouw talenten zijn talrijk en
sieren de omslag van dit boekje!
Het spreekt vanzelf dat er zonder ouders geen promovendus is… Mam, vanaf heel
vroeg heb je mij gesteund als ik een beslissing nam die wellicht niet de meest logische
was. Uiteindelijk heeft dat dus goed uitgepakt en zelfs geleid tot een ietwat exact
proefschrift. Ik prent het in voor later in de hoop dat ik ook zo mijn kinderen steun als
het nodig is, je weet maar nooit.
Pa, de appel valt niet ver van de boom, maar soms wel in een andere weide: nu jij met
pensioen bent is het voor mij wat makkelijker onze connectie toe te geven aan anderen
met als gevolg dat ik ook meer met jou bespreek. Deze Soeters-Soeters-meetings werken
inzichtgevend en bevrijdend. Leonce, jouw gastvrijheid en culinaire overdaad leiden er
telkens weer toe dat ik op dieet moet…
Pauline, Ben en Opa Han, jullie hebben mij vanaf het begin (19 jaar jong…) onvoorwaardelijk
omarmd. Het is goed dat wij elkaar zo vaak zien.
Lieve Ans, liefste Sonja, soms weet ik niet of ik jou of Hanny aan de telefoon heb; en
terwijl jullie dan lachen blijf ik vertwijfeld aan de andere kant van de lijn… Op meesterlijke
wijze haal je met regelmaat de stormwind uit de zeilen van onze viermaster! Wij houden
heel veel van je en kijken steeds weer naar je uit.
proefschrift Soeters.indb 170 4-8-2008 9:13:29
171
Chapter 10
Dankw
oord
Al heel vroeg wist ik wie mijn paranimfen moesten zijn. Peter-Jan, jij bent niet alleen
mijn oudste broer, maar vooral iemand om trots op te zijn. Jouw sociale vermogen is
benijdenswaardig, hoewel je dat zelf niet altijd door hebt… Gebruik dat, want het maakt
je sterk. “Tafel met gat” was jouw idee, het is heel mooi geworden!
Emile, wij kennen elkaar sinds de brugklas, maar werden pas echte vrienden in
Amsterdam. Samen hebben wij momenten van totale verbondenheid, veelal in het
hooggebergte, zoals tijdens onze legendarische rit van Sestriëre naar Barcelonette. De
schermutselingen begonnen vroeg die dag op de Col de l’Izoard, maar op de Col de Vars
moest ik zeldzaam diep gaan om niet uit je wiel te worden gereden. Bij die herinnering
krijg ik nog steeds spontaan zuurbranden. Volgend jaar eindelijk de Pyreneeën: afgetraind
ét pas de transpiration!
Lieve Mijntje, Diever, Voske en Pepijn; twee meisjes en twee jongens, twee paar bruine
en twee paar blauwe oogjes, alle vier verschillend met meteen al eigen meningen! Als ik
jullie ’s avonds kus voordat ik naar bed ga, droom ik wel eens dat het altijd zo mag blijven.
Maar de wetenschap dat jullie de toekomst zijn van onze wereld, doet mij accepteren dat
jullie zo snel groot worden. Ik zal er altijd voor jullie zijn.
Liefste; jij vormt het begin en het einde van dit proefschrift. Zo begint en eindigt een
dag vol geluk ook altijd met jou. Ik wil niets anders!
proefschrift Soeters.indb 171 4-8-2008 9:13:29
172
Nederlandse Biografie
Maarten René Soeters werd geboren tijdens een hittegolf op 2 augustus 1975 in
Boston, Massachusetts (U.S.A). In 1993 behaalde hij zijn Gymnasium β diploma aan het
Jeanne d’Arc College in Maastricht. Hierna werd hij in een keer ingeloot voor de studie
Geneeskunde aan de Universiteit van Amsterdam.
Naast het roeien bij de A.S.R. Nereus en wielrennen bij de A.S.C. Olympia ontving
hij zijn artsenbul op 21 december 2001. Ondanks de verleiding van het chirurgische
vak werd hij AGNIO Interne Geneeskunde in het Kennemer Gasthuis te Haarlem. Vier
maanden later werd hij aangenomen voor de opleiding Interne Geneeskunde aan het
Academisch Medisch Centrum in Amsterdam.
In het derde jaar van de opleiding (2004) werd de wetenschappelijke interesse definitief
aangewakkerd door de Metabole Groep van Prof. dr. H.P. Sauerwein, nu Dr. M.J. Serlie. Dit
resulteerde in een promotietraject waarin de opleiding Interne Geneeskunde gedurende
twee jaar werd onderbroken. Op 1 april 2008 is hij begonnen aan het aandachtsgebied
Endocrinologie en Metabolisme onder leiding van Prof. dr. E. Fliers.
Hij is getrouwd met Hanny Hamringa en samen hebben zij Mijntje, Diever, Voske en
Pepijn.
proefschrift Soeters.indb 172 4-8-2008 9:13:29