Acute and chronic suppression of the central ghrelin
signaling system reveals a role in food
Linda A.W. Verhagena,1,2, Emil Egecioglub,2, Mieneke C.M. Luijendijka,
Jacquelien J.G. Hillebrandc,d, Roger A.H. Adana,⁎,3, Suzanne L. Dicksonb,3
aRudolf Magnus Institute of Neuroscience, Department of Neuroscience and Pharmacology,
University Medical Center Utrecht, Utrecht, The Netherlands
bInstitute of Neuroscience and Physiology, Department of Physiology/Endocrinology, The Sahlgrenska Academy,
University of Gothenburg, Gothenburg, Sweden
cRintveld Centre for Eating Disorders, Altrecht Mental Health Institute, Zeist, The Netherlands
dETH Zurich, Institute of Animal Sciences, Physiology and Behaviour, Schwerzenbach, Switzerland
Received 22 February 2010; received in revised form 19 May 2010; accepted 12 June 2010
Food anticipatory activity;
Running wheel activity;
Using the rodent activity-based anorexia (ABA) model that mimics clinical features of anorexia
nervosa that include food restriction-induced hyperlocomotion, we found that plasma ghrelin
levels are highly associated with food anticipatory behaviour, measured by running wheel
activity in rats. Furthermore, we showed that ghrelin receptor (GHS-R1A) knockout mice do not
anticipate food when exposed to the ABA model, unlike their wild type littermate controls.
Likewise, food anticipatory activity in the ABA model was suppressed by a GHS-R1A antagonist
administered either by acute central (ICV) injection to rats or by chronic peripheral treatment to
mice. Interestingly, the GHS-R1A antagonist did not alter food intake in any of these models.
Therefore, we hypothesize that suppression of the central ghrelin signaling system via GHS-R1A
provides an interesting therapeutic target to treat hyperactivity in patients suffering from
© 2010 Published by Elsevier B.V.
⁎ Corresponding author. Rudolf Magnus Institute of Neuroscience, Department of Neuroscience and Pharmacology, Universiteitsweg 100,
3584 CG Utrecht, The Netherlands. Tel.: +31 88 756 8517; fax: +31 88 756 9032.
E-mail address: R.A.H.Adan@umcutrecht.nl (R.A.H. Adan).
1Present address: Institute for Genetics, Department of Mouse Genetics and Metabolism, University of Cologne, Cologne, Germany.
2These authors contributed equally to this work.
3These are both senior authors.
0924-977X/$ - see front matter © 2010 Published by Elsevier B.V.
European Neuropsychopharmacology (2011) 21, 384–392
Since its discovery in 1999, the peptide ghrelin has emerged
as an important gut–brain signal in the control of energy
balance (Hosoda et al., 2002; Kojima et al., 1999). The
ghrelin receptor (growth hormone secretagogue receptor 1A,
GHS-R1A) is highly expressed in the hypothalamus, in
particular the arcuate nucleus (Guan et al., 1997). Caloric
restriction increases ghrelin secretion, and subsequent
activation of the central ghrelin signaling system via GHS-
R1A in the arcuate nucleus has been implicated in the
stimulation of food intake (Hosoda et al., 2002). In contrast
tomany endocrine signals, plasma ghrelin levels are elevated
prior to meal initiation, decreasing once again during the
post-prandial period (Cummings et al., 2001). In addition,
acute central or peripheral ghrelin injection stimulates food
intake in rats (Horvath et al., 2001; Naleid et al., 2005).
Peripheral ghrelin injection also induces appetite in healthy
subjects (Wren et al., 2001). Furthermore, ghrelin has been
implicated in the response to long term changes in body
weight (Cummings, 2006). Collectively, these data are
indicative of a physiological role for ghrelin in hunger and
Circulating total ghrelin levels are decreased in obese
2002; Soriano-Guillen et al., 2004). Conversely, total ghrelin
can be restored by weight regain (Janas-Kozik et al., 2007;
Otto et al., 2001; Soriano-Guillen et al., 2004). In patients
suffering from anorexia nervosa, total plasma ghrelin concen-
trations of underweight subjects are increased and tend to
normalize with the recovery of body weight(Otto etal., 2001,
2005). It has been reported by Misra and colleagues that high
plasma levels of total ghrelin induced by fasting in humans are
negatively associated with the percentage body fat and with
low levels of leptin and insulin, which might play an important
role in the pathophysiology of anorexia nervosa (Misra et al.,
Activity-based anorexia (ABA) is an animal model of
anorexia nervosa, mimicking important characteristics of this
disease that include, in particular, increased locomotor
activity and reduced food consumption (Routtenberg and
Kuznesof, 1967), together with similar endocrine abnormali-
access to a running wheel and fed once per day for a limited
period of time (1–2 h). Exposure to the ABA model leads to a
chronic catabolic state caused by reduced food intake and
increased running wheel activity. Increased activity, especial-
ly prior to food intake (when ghrelin levels are presumably
high), is a hallmark feature of ABA and similar to hyperactivity
observed in patients suffering from anorexia nervosa. Inter-
estingly, acute peripheral or central ghrelin administration to
ad libitum fed mice stimulates locomotor activity (Jerlhag,
contribute to hyperlocomotor activity.
Recent studies in GHS-R1A KO mice have implicated the
central ghrelin signaling system in food anticipatory activity
(Blum et al., 2009; LeSauter et al., 2009). Therefore we first
investigated whether changes in plasma ghrelin levels were
associated with the development of hyperactivity and/or
feeding responses in the ABA model.Inspiredbythe possibility
to use GHS-R1A antagonists to suppress anorexia-induced
hyperlocomotor activity, we sought to determine whether
food anticipatory activity is suppressed in rodents adminis-
tered a GHS-R1A antagonist, either by acute central injection
the hyperlocomotor response more directly to GHS-R1A
signaling by eliminating the possibility that caloric restriction
provides the primary drive for the food anticipatory response,
incorporating studies in both GHS-R1A KO mice and GHS-R1A
2. Materials and methods
Female outbred Wistar WU rats (n=77) (Harlan, Horst, The
Netherlands), weighing 155–165 g upon arrival, were used to study
the correlation between plasma ghrelin levels and hyperactivity and
to determine the effect of central GHS-R1A antagonist treatment.
Twelve week old female C57Bl/6 mice (n=24) (Harlan, Horst, The
Netherlands) were used for studies of peripheral GHS-R1A antagonist
treatment. Heterozygous mice for the GHS-R1A disruption on a
C57Bl/6 background were provided by AstraZeneca (originally
obtained from Deltagen, San Carlos, CA, USA) and bred in house to
generate female the GHS-R1A KO and wild type littermate control
mice used in the study (Egecioglu et al., 2010). All animals were
individually housed in an ambient temperature- and humidity-
controlled room (21 °C±2 °C) under a 12-hour dark–light cycle, lights
off at 2 pm. The animals were allowed to acclimatize to the animal
facilities for one week prior to any treatment or experiment. All
procedures in rats were approved by the ethical committee on the
use and care of animals of the University of Utrecht (The Netherlands)
and all studies in mice by the local ethics committee for animal
experimentation in Gothenburg. For ethical reasons, it was decided
that the animals had to be removed from the experiment when
their body temperature was lower than 33 °C before feeding or
when the animals lost more than 25% of their initial body weight.
2.2. Measurement of plasma ghrelin in rats exposed to the
Rats were individually housed and approximately half of them began
training in the ABA model. During a 10 day training period (day −10
to 0), the ABA group (n=29) were given free access to a running
wheel and running wheel activity was continuously registered using a
Cage Registration Program (Department Biomedical Engineering,
UMC Utrecht, The Netherlands). Until the beginning of day 0, the
ABA rats were allowed water and food ad libitum. At the beginning
of the dark phase of day 0, food was removed and animals were
placed on the scheduled feeding of 1 h per day. The control group for
the study were not given access to running wheels either prior to or
during the study and were pair-fed to the ABA group (pair-fed
sedentary, n=24). Thus, the experiment was staggered to allow pair-
fed feeding: at dark onset pair-fed rats received the average amount
of food eaten by the ABA rats the day before. The body weight of all
animals was measured just prior to the dark phase, and just before
food access in both experimental groups. Every consecutive day of
the ABA model (ABA day 0–5), 4–5 animals were sacrificed by
decapitation at the end of the light phase (see Fig. 1). Trunk blood
was collected into lithium-heparin tubes (Sarstedt, Nümbrecht,
Germany) containing 83 μmol EDTA and 1 mg aprotonin. Tubes were
collected on ice until centrifugation (20 min at 3000 rpm 4 °C).
Plasma was stored at −20 °C until assay. Brains were quickly
removed, frozen and stored at −80 °C. Plasma levels of total ghrelin
(Phoenix Pharmaceuticals, Belmont, CA, USA) were measured by
commercially available radioimmunoassay (RIA) kits according to the
385 Central ghrelin antagonism inhibits food anticipatory activity
manufacturer's protocol. From the plasma samples, 2×50 μl was
taken for measurement in duplicate.
2.3. Studies exploring the effect of pharmacological
suppression of endogenous ghrelin signaling on food
anticipatory activity in the ABA model
In order to make continuous measurements of body temperature and
locomotor activity, all rats (n=24) were implanted with transmitters
(TA10TA-F40 Data Sciences International, St. Paul, Minnesota) in the
abdominal cavity under fentanyl/fluanisone (0.1 ml/100 g body
weight, IM; Hypnorm, Janssen Pharmaceutica, Beerse, Belgium)
and midazolam (0.05 ml/100 gbody weight, IP; Dormicum, Hoffman-
LaRoche, Mijdrecht, The Netherlands) anesthesia. After surgery,
rats were treated with the anti-inflammatory agent carprofen
(0.01 ml/100 g body weight, s.c.; Rimadyl, Pfizer Animal Health,
Capelle a/d Ijssel, The Netherlands) and saline (3 ml, s.c.) and
allowed to recover for at least two weeks. In order to investigate the
effects of pharmacological suppression of endogenous ghrelin on
activity behaviour and food intake, a GHS-R1A antagonist was
administered via a chronically implanted ICV catheter. At day −10,
animals were anesthetized (as indicated above) and provided with
an ICV cannula placed into the lateral ventricle; coordinates were
1.0 mm posterior from bregma, 1.0 mm lateral from midline, 5 mm
below the surface of the brain, and finally fixed in place with two
small screws and dental cement. After recovery, animals were
individually housed in cages with running wheels for a training period
of ten days (from day −10 to day 0). Running wheel activity was
continuously registered. Until the beginning of day 0, animals were
allowed food and water ad libitum. Animals were divided into
experimental groups, matched for body weight and baseline running
wheel activity. Baseline running wheel activity was determined as
the average running wheel activity during four days prior to the start
of scheduled feeding (day −8 to day −4). Body weight of all animals
was measured just prior to the one hour period of food access at the
beginning of the dark phase (see Fig. 1). At the end of day −4 and day
4 of the experiment, just before animals normally display their daily
hyperactivity, rats received an ICV injection of a GHS-R1A antagonist
(concentration 4, 8 or 16 μg) or an equal volume (3 μl) of saline
vehicle. The GHS-R1A antagonist was kindly provided by Æterna-
Zentaris GmbH (JMV2959, Frankfurt am Main, Germany). At the end
of day 5, animals were sacrificed by decapitation as described
One day prior to the start of habituation to the running wheel
setup female C57Bl/6 mice were anaesthetized with isofluorane and
surgically implanted with a subcutaneous osmotic pump (model
1002, 12 day pump from Alzet, DURECT, Cupertino California)
delivering either vehicle (saline) or JMV2959 (6 or 12 mg/kg/day)
for the duration of the experiment (9 days total). The surgical
procedure lasted for approximately 5 min. The following day the
animals were individually housed in cages with running wheels for a
training period of 5 days (from day −5 to day 0) with ad libitum
access to chow and water. On day 0, food was removed 2 h into the
dark phase and the animals placed on scheduled feeding for 2 h each
day from the onset of the dark phase. Running wheel activity was
monitored continuously, food intake was monitored daily and body
weights were recorded before the onset of the dark phase. The
animals were killed following 3 days on the scheduled feeding.
2.4. Studies exploring the effect of genetic knockout of
the ghrelin receptor, GHS-R1A, on food anticipatory
activity in the ABA model
Female GHS-R1A KO and wild type littermate mice were individually
housed in cages with running wheels for a training period of 5 days
(from day −5 to day 0) with ad libitum access to chow and water. On
day 0, food was removed 2 h into the dark phase and the animals
were placed on scheduled feeding for 2 h each day from the onset of
the dark phase. As differences in anorexic response (fatigue)
secondary to changes in caloric intake may itself induce divergent
food anticipatory activity responses, and given that young female
GHS-R1A KO mice are not able to adapt their food intake on a 4 h
feeding schedule (Blum et al., 2009) a third experimental group was
introduced, namely female GHS-R1A KO mice pair-fed to eat the
same amount of food as the wild type mice. In this group, pair
feeding was made possible by removing the time restriction and they
were allowed to eat an identical amount of food as the wild type
animals. Notably, all the food was consumed within approximately
2.5 h in this group. Of interest, 2 h and 3 h but not 4 h food
restriction in female C57Bl/6 mice of the approximate same age as
the mice used here induces a robust FAA and anorexic response on
day 1–3 of scheduled feeding in the ABA model (Kas et al., 2004).
Food intake was monitored daily and body weights were recorded
before the onset of the dark phase. Due to limitations within the
ethical permissions regarding maximum body weight loss, the
experiment was ended following 3 days of scheduled feeding.
2.5. Localization of injections
Series of 16 μm coronal sections of the brain were sliced using a
cryostat (Leica, Rijswijk, The Netherlands), thaw-mounted onto
RNAse free Superfrost slides (Menzel, Germany) and stored at −80 °C
until processing. For the localization of ICV injections, brain sections
(16 μm) were stained with Cresyl violet. Cannula placement was
defined appropriate when positioned in the lateral ventricle, based
on Paxinos brain atlas (Paxinos and Watson, 1998). Rats with
incorrect cannula placements were removed from further analysis.
2.6. Statistical analysis
All data are expressed as mean±standard error. In order to measure
plasma ghrelin levels during the course of the ABA model, groups of
rats were decapitated at consecutive days. Therefore, average daily
varying number of individuals in the running group (varying from 29
till 4 rats) and pair-fed group (varying from 24 till 4 rats) over time.
On day 4 and 5 of this experiment, association between physiological
light phase. At day 0, food was removed for the restricted experimental groups and placed on scheduled feeding of 1–2 h at the
beginning of the dark phase. Food intake and body weight were measured daily.
Experimental set-up for the ABA model. Schema of experimental set-up. Filled bars indicate the dark phase, open bars the
386 L.A.W. Verhagen et al.
parameters and the development of hyperactivity in the light phase
was investigated using Pearson's bivariate correlation analysis. We
analyzed possible associations between plasma ghrelin levels and
food anticipatory activity on day 4 and 5 while on these days
hyperactivity developed in the light phase. All data were analyzed
using SPSS 11.5 for Windows, using ANOVA with Bonferroni post hoc
testing or Student's t-tests when appropriate. Statistical significance
was set at p≤0.05.
3.1. Physiological parameters during the
development of ABA in rats
Baseline body weight during ad libitum feeding was not
significantly different between rats in running wheel cages
and sedentary rats (213.7±1.4 g versus 218.9±2.8 g respec-
tively). Running rats had a higher baseline food intake as
compared to sedentary rats (19.7±0.3 g versus 18.1±0.4 g
respectively, p=0.001), that likely reflects a compensation for
the increased physical activity. During food restriction, body
weight decreased in both ABA rats and pair-fed sedentary rats.
ABA rats had a lower relative body weight compared to pair-fed
sedentary rats (day 1 to day 4, pb0.001; day 5, p=0.008;
Fig. 2). Just prior to day 5, the body weight of the ABA rats
decreased remarkably (Fig. 2).
3.2. Endocrine parameters during the development
of ABA in rats
over time (F(5,28)=4.50, p=0.01) whereas no significant changes
were observed for plasma ghrelin levels in pair-fed sedentary rats
(F(5,23)=2.12, Table 1).
3.3. Association between plasma ghrelin levels and
hyperactivity in rats
Our food anticipation studies are built upon the well-documen-
ted observation that animals become more active during hours
preceding food access. Based on a review by Mistlberger
(Mistlberger, 1994), we defined food anticipatory activity as
the running wheel activity during the four hours prior to food
access on the days of scheduled food access. Since, under
restriction conditions food was given during the first hour of the
dark phase, the running wheel activity in the last four hours of
the light phase was taken into consideration for the evaluation
of food anticipatory activity.
feature of ABA and comparable to hyperactivity observed in
patients sufferingfromanorexianervosa (Hillebrandetal.,2008).
We therefore sought to determine whether high plasma ghrelin
from ABA day 4 and day 5 were combined, corresponding to a
period when food anticipatory activity is usually fully expressed.
Correlation analysis on day 4 and 5 revealed that plasma ghrelin
levelswerepositivelycorrelatedwith food anticipatoryactivity in
ABA rats (r=0.760, p=0.018, Fig. 3.A.). There was no correlation
between plasma ghrelin levels and total running wheel activity on
the same days (Fig. 3.B.).
3.4. Effect of acute central injection of a GHS-R1A
antagonist on food anticipatory activity in rats
To investigate whether central ghrelin signaling contributes to
the development of hyperactivity caused by caloric restriction,
we injected a ghrelin receptor (GHS-R1A) antagonist centrally
into the brain ventricles during ad libitum fed and food-
restricted conditions. As already indicated, the running wheel
activity in the last four hours of the light phase was taken into
consideration for the evaluation of food anticipatory activity.
To circumvent the large variation in running wheel activity
between rats within the same experimental group, food
anticipatory activity is presented as percentage of total running
wheel activity during the same day of acute ICV injection with
the GHS-R1A antagonist or saline vehicle. The %food antici-
patory activity after ICV injection at the end of day 4 was
compared to the %food anticipatory activity on the day before
sedentary rats. Daily body weight (as percentage of initial
body weight, start of food restriction) in food-restricted running
rats fed for 1 h (black circle) and pair-fed sedentary rats (open
circle). Significant differences between pair-fed sedentary and
1-hour fed running (ABA) rats are indicated by asterisk;
Body weight changes in ABA rats and pair-fed
Table 1 Plasma ghrelin levels during scheduled feeding.
Plasma ghrelin levels (ng/ml)
Day 1-hour fed running ratsPair-fed sedentary rats
Total plasma ghrelin levels in 1-hour fed running rats (ABA) and
pair-fed sedentary rats. Significant differences within the
experimental groups compared to day 0 are indicated by “a”,
significant different from ABA rats at the same day are indicated
by “b”. ANOVA, Bonferroni, with significance set at pb0.05.
387Central ghrelin antagonism inhibits food anticipatory activity
(the end of ABA day 3). Acute injection with the highest
concentration of the GHS-R1A antagonist (JMV2959 16 μg) in the
lateral ventricle tended to decrease food anticipatory activity
(p=0.06) whereas the lower concentrations had no effect
3.5. Effect of acute central injection of a GHS-R1A
antagonist on food intake in rats
In ad libitum fed rats, acute ICV injection of the GHS-R1A
antagonist (16 μg) reduced daily food intake (Fig. 5.A.). A
tendency towards a decrease in food intake was observed using
a slightly lower dose (8 μg) of the antagonist. Vehicle injection
or injection with the lowest dose of the antagonist (4 μg) did not
affect daily food intake. In ABA rats, acute ICV injection of the
GHS-R1A antagonist had no effect on daily (restricted) food
intake (Fig. 5.B.). Thus, acute ICV injection of a GHS-R1A
antagonist reduces food intake in ad libitum fed running rats,
but not in ABA rats.
3.6. Lack of food anticipatory activity in ghrelin
receptor KO mice
To confirm the involvement of ghrelin signaling in food
anticipatory activity found in rats, we exposed wild type and
GHS-R1A KO mice to the ABA model. Since scheduled-fed GHS-
R1A KO mice exhibited a reduced daily food intake as compared
to the control wild type mice during the scheduled feeding
(total 3 day food intake; wild type: 5.5±0.4 g, scheduled-fed
GHS-R1A KO: 3.6±0.5 g, pb0.001, Student's t-test), we in-
cluded an additional experimental group: GHS-R1A KO mice that
were pair-fed to the wild type mice. Food intake was not
significantly different between the groups during the 5 day
Over the whole period of scheduled food restriction, pair-fed
GHS-R1A KO mice lost less body weight as compared to wild type
and the scheduled-fed GHS-R1A KO mice (pb0.01, ANOVA
followed by Bonferroni post hoc test, Fig. 6.B.).
Because of severe body weight loss and the significant
differences observed in body weight loss on ABA day 3, we
analyzed running wheel activity levels on day 2 of the ABA
model in wild type and GHS-R1A KO mice. Baseline running
wheel activity was not significantly different between the
levels and (A) total food anticipatory activity (food anticipatory activity (FAA), measured by wheel revolutions) or (B) total running
wheel activity levels in running rats exposed to 1 h feeding schedule (ABA, n=9). Statistical significance was set at pb0.05.
Correlation between ghrelin levels and food anticipation. Correlation analysis on day 4 and 5 between plasma ghrelin
tion. Effect of acute ICV injection of saline (n=6) or the GHS-R1A
antagonist, JMV2959, on food anticipatory activity in rats
exposed to the ABA model. On day 4, JMV2959 was administered
at 3 different doses: 4 μg (n=10), 8 μg (n=6) and 16 μg (n=6).
The percentage food anticipatory activity was defined as the
running wheel activity 4 h prior to food access as percentage of
total running wheel activity on days of scheduled feeding. The
white bars represent the mean percentage food anticipatory
activity on the day before ICV injection (ABA day 3). The black
bars represent the mean percentage food anticipatory activity
after ICV injection (ABA day 4). A tendency towards a reduction
in percentage food anticipatory activity was observed in rats
with an acute ICV injection with JMV2959 16 μg (p=0.06).
Acute effect of GHS-R1A antagonist on food anticipa-
388 L.A.W. Verhagen et al.
groups during the 5 days habituation. During the course of the
ABA model, running wheel activity in the dark phase remained
similar between wild type, schedule-fed GHS-R1A KO, and pair-
fed GHS-R1A KO groups. Wild type mice clearly developed food
anticipatory activity (Fig. 6.A.). Scheduled-fed GHS-R1A KO also
developed a food anticipatory response however this was
attenuated compared to wild type mice (pb0.05, Fig. 6.A.).
Day time running wheel activity of pair-fed GHS-R1A KO mice on
day 2 was significantly decreased relative to both wild type and
schedule-fed GHS-R1A mice (both pb0.05) and did not differ
from the habituation when the animals had ad libitum access to
food. Thus, lack of ghrelin signaling in GHS-R1A KO mice results
in reduced anticipation to food.
3.7. Effects of chronic pharmacological suppression of
central ghrelin signaling on food anticipatory activity
To extend our acute studies showing that a GHS-R1A antagonist
suppresses food anticipatory activity in the ABA model, an
additional study was undertaken in which food anticipatory
activity was measured in ABA mice following chronic peripheral
administered using three different doses just prior to day 5 on 24 h food intake: 4 μg (ad libitum fed n=6; food-restricted n=10), 8 μg
(ad libitum fed n=10; food-restricted n=6), or 16 μg (ad libitum fed n=8; food-restricted n=6). (A) ad libitum fed running rats and
(B) rats exposed to the ABA model. Food intake is presented in absolute values. Significant differences are indicated by asterisks,
*pb0.05, Student's t-test.
Acute effect of GHS-R1A antagonist on food intake. Effect of an acute ICV injection of saline (n=6) or a GHS-R1A antagonist
activity (A) and relative body weights (B) in wild type mice (white circle, n=6), scheduled-fed GHS-R1A KO mice (black circle, n=6),
and pair-fed GHS-R1A KO mice (gray circle, n=7) exposed to the ABA model. Black bars indicate dark phase, and white bars
indicate light phase. Hourly running wheel activity is presented as mean±SEM, and body weights are plotted as a percentage of
initial body weight. Differences in running wheel activity during the dark phase or light phase was analyzed using ANOVA repeated
measures followed by Bonferroni post hoc test. Significant differences between wild type and schedule-fed GHS-R1A KO mice are
indicated by “a”, whereas “b” indicates significant differences between wild type and pair-fed GHS-R1A KO mice. Statistical
significance was set at pb0.05.
Activity levels and body weight changes in ghrelin receptor (GHS-R1A) KO mice exposed to the ABA model. Running wheel
389 Central ghrelin antagonism inhibits food anticipatory activity
administration of the same antagonist at two different doses,
namely 6 and 12 mg/kg/day of JMV2959. Baseline running wheel
activity (during habituation) did not differ between the JMV2959-
and saline-treated groups. Similar to the analysis of food
anticipatory activity in the GHS-R1A KO study, we analyzed food
anticipatoryactivity on ABA day 2 (Fig. 7). The highest dose of the
antagonist (12 mg/kg/day), reduced food anticipatory activity
weight and food intake were unaffected by the GHS-R1A
antagonist compared to vehicle treatment regardless of the dose
(6 mg/kg/day dose: BW, saline: 15.6±0.3 g JMV2959: 15.9±0.4,
food intake, saline: 0.65±0.2 g JMV2959: 0.49±0.7 g, 12 mg/kg/
saline: 2.5±0.2 g JMV2959: 2.3±0.5 g). Thus, continuous periph-
eral administration of the GHS-R1A antagonist suppressed food
anticipation without affecting the regulation of body weight or
In the present study we demonstrate that the central ghrelin
signaling system, involving GHS-R1A, is required food
anticipation, measured in the ABA model. Thus, food
anticipatory activity was attenuated both in chronic models
of suppressed ghrelin signaling (GHS-R1A KO mice and mice
chronically treated with a GHS-R1A antagonist) and in an
acute model (rats given an acute central injection of the
GHS-R1A antagonist). Importantly, the effect on food
anticipatory activity was independent of food intake in
both the acute and the chronic studies using the GHS-R1A
antagonist. Collectively these data suggest that the gut–
brain signal provided by ghrelin is important for food
anticipation and that GHS-R1A antagonists provide a poten-
tial therapy for suppressing food anticipatory hyperlocomo-
tor activity independently of food intake.
Our data, showing suppressed food anticipatory activity in
GHS-R1A KO mice, are in good agreement with recent papers
on this topic (Blum et al., 2009; LeSauter et al., 2009).
Rhythmic ghrelin and the circadian clock gene PER expres-
sion in stomach oxyntic cells are synchronized to timing of
food delivery and GHS-R1A KO mice show less food
anticipatory activity when food is given for 6 h (LeSauter et
al., 2009) or for 4 h (Blum et al., 2009) in the light phase. In
the present study, we compared GHS-R1A KO and wild type
mice put on a 2 h feeding schedule in the beginning of the
dark phase which rapidly induces both a FAA and an anorexic
response when combined with free access to running wheels.
Given that young female GHS-R1A KO mice are not able to
adapt their food intake on a 4 h feeding schedule (Blum et
al., 2009) we also included a third experimental group, GHS-
R1A KO mice that were pair-fed to the wild type group (by
increasing the amount of time exposed to the food). As food
anticipatory activity was decreased in both the GHS-R1A KO
group on a 2 h feeding schedule and the pair-fed GHS-R1A KO
group compared to wild type mice, it seems likely that the
suppressed food anticipatory activity (i.e. hyperlocomotor
activity) is not secondary to the caloric deficit and body
weight loss, but rather, relates directly to suppressed GHS-
R1A signaling. Furthermore, night time running activity was
also similar between all three groups indicating that the
changes in FAA found was not due to fatigue induced by
differences in caloric intake. However, the decrease in FAA
in the pair-fed GHS-R1A KO group may be a consequence of
the longer exposure to food per se rather than the lack of
ghrelin signaling. Conversely, robust FAA has been reported
on day 1–3 following the start of food restriction in mice of
the same strain and age using either 2 h or 3 h scheduled
feeding protocols (Kas et al., 2004).
Previously we reported that central administration of the
ghrelin antagonist JMV2959 suppresses ghrelin-induced food
intake (Salome et al., 2009a) and ghrelin-induced adiposity
(Salome et al., 2009b). However, in the present study we
were unable to detect any effects of the antagonist on food
intake per se, in the dose range used to suppress food
anticipatory response. Thus, we found that the highest dose
wheel activity on day 2 of the ABA model in mice (A) treated with JMV2959 6 mg/kg/day (black circle, n=6) and (B) treated with
JMV2959 12 mg/kg/day (black circle, n=6) as compared to mice treated with saline solution (white circle, n=6 and n=5 respectively).
Filled black bars indicate the dark phase, open bars the light phase. Differences in running wheel activity during the dark phase or light
phase was analyzed using ANOVA repeated measures followed by Bonferroni post hoc testing. Significant differences between wild
type and GHS-R1A antagonist-treated mice are indicated by “a”. Statistical significance was set at pb0.05.
The effect of continuous peripheral administration of GHS-R1A antagonist on activity levels in ABA mice. Hourly running
390 L.A.W. Verhagen et al.
of the GHS-R1A antagonist tested (16 μg JMV2959, ICV)
reduced food intake in ad libitum fed running rats but had no
effect on food intake in the ABA group. Similar effects on
food anticipatory activity were also observed in rats upon
acute administration of the antagonist as well as in a chronic
study involving peripheral treatment of the same antagonist
(12 mg/kg/day) to mice exposed to the ABA model. Given
that plasma ghrelin levels are associated with food antici-
patory activity but not with total running wheel activity and
also that the GHS-R1A antagonist suppresses food anticipa-
tory activity, our data provide strong evidence that the
central ghrelin signaling system (via GHS-R1A) is involved in
food anticipation. This hypothesis is further supported by the
aforementioned experiments in which we show a lack of food
anticipatory activity in the GHS-R1A KO mice exposed to the
ABA paradigm. We may infer that GHS-R1A is not only
involved in the development of food anticipatory activity
(from studies in GHS-R1A KO mice) but that it is possible to
manipulate food anticipatory activity using acute therapeu-
tic intervention involving GHS-R1A antagonists.
In the present study we also demonstrate an association
between plasma ghrelin levels and food anticipatory activity
in the ABA model. Thus, in ABA rats there was a strong
positive correlation between the expression of food antici-
patory activity (i.e. running wheel activity during the four
hours prior to scheduled food access) and plasma ghrelin
levels on day 4 and day 5 of exposure to the ABA paradigm.
Interestingly, plasma ghrelin levels were not correlated with
total (24 h) running wheel activity but only to the food
anticipatory phase, supporting the idea that ghrelin's
locomotor stimulatory activity may be especially linked to
the food anticipatory period.
Studies employing Fos activity in GHS-R1A KOs have
highlighted key hypothalamic areas that are likely to be of
importance for ghrelin's effects on food anticipatory activity
(Blum et al., 2009). Given that ghrelin also targets the
midbrain dopamine system to induce locomotor activity
(Jerlhag et al., 2006; Jerlhag et al., 2007), further studies
are required to discover whether this system is also
important for ghrelin's role in food anticipatory (hyperloco-
motor) activity. Moreover, it will be interesting to discover
whether increased anticipatory activity contributes to
ghrelin's effects to promote reward-seeking behaviour, not
only for food but also for chemical drugs such as alcohol
(Jerlhag et al., 2009).
Anorectic patients often display abnormally high physical
activity levels (Kron et al., 1978), which hinders the process
of recovery (Holtkamp et al., 2004; Kaye et al., 1988). Thus,
reducing hyperactivity in severely ill patients suffering from
anorexia nervosa could be beneficial for therapeutic
outcome. The present data demonstrating that suppression
of the central ghrelin signaling system (incorporating both
pharmacological and genetic models) reduces hyperactivity
without influencing food intake in the ABA paradigm suggests
that the ghrelin receptor, GHS-R1A, could provide a
therapeutically relevant target for treatment of anorexia
nervosa. Interestingly, ghrelin infusion did not increase
appetite in patients suffering from anorexia nervosa as
compared to weight matched healthy controls (Miljic et al.,
2006). In conclusion, our data support the hypothesis that
suppression of the central ghrelin signaling system could
suppress hyperactive behaviour in patients suffering from
anorexia nervosa, based on rodent studies incorporating the
Role of the funding source
L.A.W. Verhagen was supported by NWO grant no. 90339175, The
Netherlands. NWO had no further role in study design; in the
collection, analysis and interpretation of data; in the writing of the
report; and in the decision to submit the paper for publication.
Author LAWV, EE, and JJGH designed the study, wrote the protocol
and undertook the statistical analysis. The contribution of author
MCML enclosed practical assistance. Authors RAHA, SLD, LAWV and
EE wrote the manuscript. All authors contributed to and have
approved the final manuscript.
Conflict of interest
None of the authors has conflict of interests associated with the work
reported in this paper.
The authors are indebted to Æterna Zentaris GmbH (Germany) for
providing the GHS-R1A antagonist JMV2959 used in this study.
Furthermore, we would like to thank Mohammad Bohlooly-Y from
Astra Zeneca R&D (Mölndal, Sweden) for providing the GHS-R1A KO
mice. LAWV was supported by the Netherlands Organisation for
Scientific Research (NWO grant no. 90339175), The Netherlands. SLD
and RAHA received support from the EU (FP7-HEALTH-2009-241592,
FP7-KBBE-2009-3-245009). SLD also received support from the
Swedish Medical Research Council (K2007-54X-20328-013), ALF
Göteborg (SU7601) and the Swedish Foundation for Strategic
Research to Sahlgrenska Center for Cardiovascular and Metabolic
Research (A305-188). EE was supported by Konrad & Helgfrid
Johanssons Fond and Fredrik & Ingrid Thurings Stiftelse.
Blum, I.D., Patterson, Z., Khazall, R., Lamont, E.W., Sleeman, M.W.,
Horvath, T.L., Abizaid, A., 2009. Reduced anticipatory locomotor
responses to scheduled meals in ghrelin receptor deficient mice.
Neuroscience 164, 351–359.
of appetite and body weight. Physiol. Behav. 89, 71–84.
Cummings, D.E., Purnell, J.Q., Frayo, R.S., Schmidova, K., Wisse,
B.E., Weigle, D.S., 2001. A preprandial rise in plasma ghrelin
levels suggests a role in meal initiation in humans. Diabetes 50,
de Rijke, C.E., Hillebrand, J.J., Verhagen, L.A., Roeling, T.A., Adan,
R.A., 2005. Hypothalamic neuropeptide expression following
chronic food restriction in sedentary and wheel-running rats. J.
Mol. Endocrinol. 35, 381–390.
Egecioglu, E., Jerlhag, E., Salomé, N., Skibicka, P.K., Haage, D.,
Bohlooly-Y., M., et al., 2010. Ghrelin increases intake of
rewarding food in rodents. Addict. Biol. 15, 304–311.
Guan, X.M., Yu, H., Palyha, O.C., McKee, K.K., Feighner, S.D.,
Sirinathsinghji, D.J., et al., 1997. Distribution of mRNA encoding
the growth hormone secretagogue receptor in brain and
peripheral tissues. Brain Res. Mol. Brain Res. 48, 23–29.
391Central ghrelin antagonism inhibits food anticipatory activity
Hansen, T.K., Dall, R., Hosoda, H., Kojima, M., Kangawa, K., Download full-text
Christiansen, J.S., Jorgensen, J.O., 2002. Weight loss increases
circulating levels of ghrelin in human obesity. Clin. Endocrinol.
(Oxf.) 56, 203–206.
Hillebrand, J.J., Kas, M.J., van Elburg, A.A., Hoek, H.W., Adan, R.A.,
2008. Leptin's effect on hyperactivity: potential downstream
effector mechanisms. Physiol. Behav. 94, 689–695.
Holtkamp, K., Hebebrand, J., Herpertz-Dahlmann, B., 2004. The
contribution of anxiety and food restriction on physical activity
levels in acute anorexia nervosa. Int. J. Eat. Disord. 36, 163–171.
Horvath, T.L., Diano, S., Sotonyi, P., Heiman, M., Tschöp, M., 2001.
Minireview: ghrelin and the regulation of energy balance—a
hypothalamic perspective. Endocrinology 142, 4163–4169.
Hosoda, H., Kojima, M., Kangawa, K., 2002. Ghrelin and the
regulation of food intake and energy balance. Mol. Interv. 2,
Janas-Kozik, M., Krupka-Matuszczyk, I., Malinowska-Kolodziej, I.,
Lewin-Kowalik, J., 2007. Total ghrelin plasma level in patients
with the restrictive type of anorexia nervosa. Regul. Pept. 140,
Jerlhag, E., 2008. Systemic administration of ghrelin induces
conditioned place preference and stimulates accumbal dopa-
mine. Addict. Biol. 13 (3–4), 358–363.
Jerlhag, E., Egecioglu, E., Dickson, S.L., Andersson, M., Svensson,
L., Engel, J.A., 2006. Ghrelin stimulates locomotor activity and
accumbal dopamine-overflow via central cholinergic systems in
mice: implications for its involvement in brain reward. Addict.
Biol. 11, 45–54.
J.A., 2007. Ghrelin administration into tegmental areas stimulates
locomotor activity and increases extracellular concentration of
dopamine in the nucleus accumbens. Addict. Biol. 12, 6–16.
Jerlhag, E., Egecioglu, E., Landgren, S., Salome, N., Heilig, M.,
Moechars, D., et al., 2009. Requirement of central ghrelin
signaling for alcohol reward. Proc. Natl. Acad. Sci. U. S. A. 106,
Kas, M.J., van den Bos, R., Baars, A.M., Lubbers, M., Lesscher, H.M.,
Hillebrand, J.J., et al., 2004. Mu-opioid receptor knockout mice
show diminished food-anticipatory activity. Eur. J. Neurosci. 20,
Kaye, W.H., Gwirtsman, H.E., Obarzanek, E., George, D.T., 1988.
Relative importance of calorie intake needed to gain weight and
level of physical activity in anorexia nervosa. Am. J. Clin. Nutr.
Kojima, M., Hosoda, H., Date, Y., Nakazato, M., Matsuo, H.,
Kangawa, K., 1999. Ghrelin is a growth-hormone-releasing
acylated peptide from stomach. Nature 402, 656–660.
Kron, L., Katz, J.L., Gorzynski, G., Weiner, H., 1978. Hyperactivity
in anorexia nervosa: a fundamental clinical feature. Compr.
Psychiatry 19, 433–440.
LeSauter, J., Hoque, N., Weintraub, M., Pfaff, D.W., Silver, R.,
2009. Stomach ghrelin-secreting cells as food-entrainable circa-
dian clocks. Proc. Natl. Acad. Sci. U. S. A. 106, 13582–13587.
Miljic, D., Pekic, S., Djurovic, M., Doknic, M., Milic, N., Casanueva,
F.F., et al., 2006. Ghrelin has partial or no effect on appetite,
growth hormone, prolactin, and cortisol release in patients with
anorexia nervosa. J. Clin. Endocrinol. Metab. 91, 1491–1495.
Misra, M., Miller, K.K., Kuo, K., Griffin, K., Stewart, V., Hunter, E.,
et al., 2005. Secretory dynamics of ghrelin in adolescent girls
with anorexia nervosa and healthy adolescents. Am. J. Physiol.
Endocrinol. Metab. 289, E347–E356.
Mistlberger, R.E., 1994. Circadian food-anticipatory activity: formal
models and physiological mechanisms. Neurosci. Biobehav. Rev.
Naleid, A.M., Grace, M.K., Cummings, D.E., Levine, A.S., 2005.
Ghrelin induces feeding in the mesolimbic reward pathway
between the ventral tegmental area and the nucleus accumbens.
Peptides 26, 2274–2279.
Otto, B., Cuntz,U., Fruehauf, E.,Wawarta, R., Folwaczny, C., Riepl,
R.L., et al., 2001. Weight gain decreases elevated plasma ghrelin
concentrations of patients with anorexia nervosa. Eur. J.
Endocrinol. 145, 669–673.
Otto, B., Tschöp, M., Fruhauf, E., Heldwein, W., Fichter, M., Otto,
C., Cuntz, U., 2005. Postprandial ghrelin release in anorectic
patients before and after weight gain. Psychoneuroendocrinol-
ogy 30, 577–581.
Paxinos, G., Watson, C., 1998. The Rat Brain in Stereotaxic
Coordinates, 4th edition. Academic Press, New York.
Routtenberg, A., Kuznesof, A.W., 1967. Self-starvation of rats living
in activity wheels on a restricted feeding schedule. J. Comp.
Physiol. Psychol. 64, 414–421.
Salome, N., Haage, D., Perrissoud, D., Moulin, A., Demange, L.,
Egecioglu, E., et al., 2009a. Anorexigenic and electrophysiolog-
ical actions of novel ghrelin receptor (GHS-R1A) antagonists in
rats. Eur. J. Pharmacol. 612, 167–173.
Salome, N., Hansson, C., Taube, M., Gustafsson-Ericson, L.,
Egecioglu, E., Karlsson-Lindahl, L., et al., 2009b. On the central
mechanism underlying ghrelin's chronic pro-obesity effects in
rats: new insights from studies exploiting a potent ghrelin
receptor antagonist. J. Neuroendocrinol. 21, 777–785.
Soriano-Guillen, L., Barrios, V., Campos-Barros, A., Argente, J.,
2004. Ghrelin levels in obesity and anorexia nervosa: effect of
weight reduction or recuperation. J. Pediatr. 144, 36–42.
Wren, A.M., Seal, L.J., Cohen, M.A., Brynes, A.E., Frost, G.S.,
Murphy, K.G., et al., 2001. Ghrelin enhances appetite and
increases food intake in humans. J. Clin. Endocrinol. Metab. 86,
392 L.A.W. Verhagen et al.