nature publishing group
Behavior and Psychology
Environmental factors such as the visual exposure to food are
claimed to influence eating behavior in humans (1). Sight of
hedonic food enhanced the desire to eat and calorie intake,
even when subjects were satiated (2,3). The specific physi-
ological mechanisms, which trigger motivation for food con-
sumption independent of metabolic energy status, are so far
The neuropeptide ghrelin, mainly secreted by the stomach
and proximal small intestine, is the major stimulus for food
intake in humans and rodents. Ghrelin acts via the ghrelin
receptor (GHS-R1a), which is particularly found in the neu-
ropeptide Y and growth hormone-releasing hormone neurons
in the hypothalamus-pituitary unit (4,5). After exogenous
ghrelin administration, feeding and body weight increase in
rodents (6,7). In normal human subjects, ghrelin administra-
tion increases self-rated appetite and calorie intake (8) and
prompts the imagination of favorite meals (9). Ghrelin levels
increase in humans before meals and decrease after meals
regardless whether meal times were fixed (10) or meals were
initiated voluntarily (11). Peripheral ghrelin secretion is sup-
posed to be controlled by cephalic mechanisms, likewise the
sympathetic neural system stimulates directly peripheral ghre-
lin secretion (12,13).
In fMRI studies, it was demonstrated that the amygdala, orbito-
frontal cortex, insula, anterior cingulated, and fusiform gyrus
contain neurons, who respond specifically to visual food stimuli
(14,15). These areas participate in a complex neural circuit that
represents food identification and estimation. The intravenous
administration of ghrelin increases the response to images with
food content in most of these above-mentioned areas (16).
A reciprocal interaction exists in the energy balance between
ghrelin and leptin and between ghrelin and insulin (11,17).
Leptin is produced in the adipose tissue, and leptin recep-
tors are found in the arcuate nucleus located within the hypo-
thalamus, the same region where ghrelin is engaged. Leptin
decreases food intake and increases energy expenditure (18).
Ghrelin inhibits insulin secretion, and ghrelin secretion is
decreased by insulin (19,20). Data are conflicting, if leptin
declines before meal initiation (11,21).
As far as we know, it is unclear whether ghrelin plasma levels
are affected by external cues such as sight of food. Since such
cues are omnipresent in nowadays society it appears impor-
tant to clarify this issue. Therefore, we examined the effect of
visual presentation of images of food on ghrelin plasma levels
in healthy, normal weight subjects. In addition, plasma con-
centrations of insulin and leptin were evaluated because they
are postulated to influence the course of ghrelin secretion. In
Ghrelin Levels Increase After Pictures
Petra Schüssler1, Michael Kluge1, Alexander Yassouridis1, Martin Dresler1, Manfred Uhr1 and Axel Steiger1
The neuropeptide ghrelin is a major signal for food intake in various species including humans. After exogenous
ghrelin administration, food intake and body weight increase in rodents. In normal human subjects, ghrelin
administration increases self-rated appetite and calorie intake and prompts the imagination of favorite meals. It is
unclear so far whether ghrelin levels are affected by external cues such as sight of food. We investigated the influence
of pictures showing food compared to neutral pictures on ghrelin levels in young normal male subjects (n = 8). The
study consisted of two consecutive sessions with a one-week interval. During each session, blood for later analysis of
plasma concentrations of ghrelin was collected between 08:15 and 13:00 every 15 min (between 10:30 and 11:30 every
10 min). Breakfast and lunch was provided at 08:30 and 12:00, respectively. Fifty pictures were presented from 10:30
to 10:45 showing neutral images during the first session and food contents during the second session. As expected,
ghrelin levels increased before each meal independent of the picture contents. In addition, ghrelin levels during the
30-min interval following the presentation of pictures with food increased significantly compared to the 30-min interval
before this presentation (area under the curve (AUC): 188 % vs. 158 %, P < 0.05). The difference in the increases
between the two picture conditions was also significant (P < 0.05). Our findings suggest that sight of food elevates
ghrelin levels in healthy volunteers.
Obesity (2012) doi:10.1038/oby.2011.385
1Max Planck Institute of Psychiatry, Munich, Germany. Correspondence: Axel Steiger (firstname.lastname@example.org)
Received 3 August 2011; accepted 1 December 2011; advance online publication 12 January 2012. doi:10.1038/oby.2011.385
Behavior and Psychology
more detail, we wanted to test the hypothesis that ghrelin lev-
els increase in healthy volunteers after presentation of pictures
showing food. The courses of leptin and insulin in this experi-
ment are difficult to predict. Since ghrelin and these hormones
may interact, in addition the course of their plasma concentra-
tions was assessed.
Methods and Procedures
The sample consisted of eight healthy male volunteers of normal weight
(mean age 22.3 ± 5.5 years, range 21–28, mean BMI 22.9 ± 0.9 kg/m2).
Detailed characteristics are given in Supplementary Table S1 online.
All subjects were drug-free for at least 3 months and entered the study
after passing rigid psychiatric, physical, and laboratory examinations.
All were emmetropic or corrected to normal vision. Reasons for exclu-
sion from the study were as follows: major or chronic diseases (e.g.,
eating disorders, diabetes, heart failure, hepatitis), acute or previous
psychiatric or chronic neurological disorder (e.g., schizophrenia, epi-
lepsy) in the own or family history, stressful life events, shift work, aber-
rancies in the blood chemistry or electrocardiogram. Other exclusion
factors were excessive physical exercise; food allergies; specific diet
(e.g., vegan, vegetarian); eating habits differing from the local culture,
abuse of drugs, nicotine (more than 5 cigarettes per day), loss or gain of
weight in the last 3 months more than 3 kg.
The experiment was approved by the Ethics Committee for Human
Experiments of the University of Munich. After the purpose of the study
had been explained to the subjects, all of them gave their informed con-
sent according to the tenets of the declaration of Helsinki.
Endocrine measurements. Plasma total ghrelin and leptin were measured
using commercial radio-immunoassays (ghrelin: Phoenix, Belmont,
CA; intra- and interassay coefficients of variation <13%; leptin: Linco
Research, Saint Charles, Mo; intra- and interassay coefficients were
below 7% and 9%, respectively). For the quantitative measurement of
insulin, a chemiluminescent immunoassay was used (Immulite 2000;
Siemens, Eschborn, Germany, intra- and interassay coefficients of vari-
ation were below 5.5% and 7.35%).
Random samples for each hormone were analyzed in duplicate.
According to standard procedures for time series, the remaining speci-
mens were analyzed only once.
Visual presentation. Fifty color-photographs of food and 50 nonedible
objects were selected for visual presentation on a computer screen (23 ×
37.5 cm). The taste intensity of the 50 food photographs was scored by
21 male volunteers, who were asked to assign to them a score between 1
(very tasty) and 7 (not tasty).
The food photographs demonstrated savory and sweet meals which are
typical for lunch in Germany (e.g., steak, Viennese Schnitzel, pizza, ice
cream, chocolate cake). The nonfood images consisted of various objects
and sceneries without any context to eating behavior (bicycle, piano, pair
of shoes, clouded sky etc.).
During the experiment, images were presented to each of the eight
subjects in randomized order for a total time of 15 min. Each picture was
shown on average three times for 6 s.
At the end of the second session, subjects were committed to evalu-
ate the picture’s taste intensity using the same scale as the 21 volunteers.
This task was necessary in order to examine the objectivity in scoring the
contents of the used pictures.
The study consisted of two consecutive sessions with a 3- to 8-day inter-
val. During each session, 5 ml blood was drawn from the adjacent room,
using an intravenous cannula and a tubic extension for later analysis of
plasma concentrations of ghrelin, leptin, and insulin every 15 min from
08:15 to 10:30, from 11:30 to 13:00, and every 10 min between 10:30
and 11:30. We choose a frequency of blood sampling which is higher
than in most other studies on the course of ghrelin levels in humans
(11,22), but in accordance to a recent study by Spiegel et al. (23). Similar
to their study, we elevated the sampling frequency during the interval of
major interest (10:30 to 11:30). Breakfast (1 cup of coffee, 2 buns, but-
ter, jam, 2 slices cheese or sausage) and lunch (meat, pasta or potatoes,
and vegetables and 2 glasses of water) were provided at 08:30 and 12:00,
respectively. Such meals are typical for Germany.
Fifty pictures in 6-s intervals were presented from 10:30 to 10:45 show-
ing neutral images during the first session and food pictures during the
second session. This order was chosen to avoid anticipation in the sub-
jects during the second session. A flow chart of the experiment is given
in Figure 1.
Subjects were not informed before of the food-related content of the
photographs presented on a computer screen from a distance around
1 m. They were informed in advance to receive breakfast and lunch dur-
ing the experiment.
All subjects reported regular eating patterns including breakfast, lunch,
and dinner. Study participants were advised to have dinner as usually and
not to be sleep deprived before the experiment.
The experiment was carried out in a quiet room, and subjects were
placed in a comfortable armchair in sitting position. Subjects were
watched by video monitoring not to fall asleep. Beside breakfast, lunch,
and the time for presentation on the computer screen, subjects were
allowed to read magazines or newspapers, which were free of any eating-
At the end of each session at 13:00, subjects completed a recognition
test, presenting five of the previously seen pictures with food or neutral
content and five new pictures with food or neutral content randomly
intermixed. Subjects rated whether each photograph had been seen previ-
ously (yes) or had not been seen (no) during the experiment. The number
of correct answers was counted for each subject.
Metrical variables are expressed as mean ± s.d. Differences of mean
ghrelin, leptin, and insulin plasma levels subsequent to presenta-
tion of neutral or food pictures at single time points were investi-
gated only on the exploratory level. For inferential statistics, the area
under the curve (AUC) determined according to the trapezoid rule
and the mean location (ML) was taken into account for each hor-
mone and each of two intervals (phases) namely the preintervention
(10:00–10:30) and postintervention period (10:50–11:20). The phase
Three- to 8-day interval
Figure 1 Flow chart of the experiment.
Behavior and Psychology
and picture- stimulus effects on the AUCs and MLs of each of the hor-
mones were then tested about significance by two-factorial multivari-
ate analyses of variance with repeated measures designs. Thereby both
influential factors “phase” and “picture stimulus” were two within-
subjects-factors with two levels. To assess better the picture-stimulus
effect on changes of the hormone levels before and after the stimulus,
we calculated in addition for each hormone the quotients of the AUC
(ML) in the postintervention phase to the AUC (ML) in the preinter-
vention phase and then applied a multivariate analysis of variance on
all of these quotients. Possible associations between the three hor-
mones at each of the three intervals were tested about significance by
means of the Pearson correlation coefficients. The Pearson correlation
coefficient was also used to test agreement between subjects and raters
in the scoring of the pictures’ intensity of taste.
Differences in the nonendocrine data (recognition test) were tested
by the nonparametric Wilcoxon matched pairs’ tests. As nominal level
of significance, α = 0.05 was accepted and corrected by all post hoc tests
according to Bonferroni procedure.
Ghrelin, insulin, and leptin
Table 1 shows some descriptive and inferential statistics concern-
ing the AUC and ML values of the hormones. Supplementary
Figure S1 online shows ghrelin, insulin, and leptin levels of each
subject from 08:15 to 13:00. ANOVA revealed for ghrelin sig-
nificant interaction effects of the factors (Wilks multivariate test
of significance; effect of phase × picture-stimulus: F(2,6) = 5.87,
sig of F = 0.039), which were significantly remarkable on both
the ML and AUC values as well (univariate F-tests, P < 0.05).
By scrutinizing the simple effects of phase and picture stimulus
on ghrelin, i.e., by investigating the phase differences in its ML
and AUC values within each picture stimulus separately and
vice versa, we identified that both ghrelin ML and AUC values
increased significantly during the 30-min interval following the
presentation of food pictures (intervention) compared to the
preintervention interval. When neutral pictures were presented,
then only ghrelin AUC values showed significant differences
between the two presentation periods (tests with contrasts in
multivariate analysis of variance (MANOVA), P < 0.05).
Leptin showed only a significant phase effect (Wilks multi-
variate test of significance; effect of phase: F(2,6) = 19.26, sig of
F = 0.002) attributed exclusively to the AUC values (univariate
F-tests, P < 0.05), whereas insulin seemed not to be affected
neither from the picture stimulus nor from the intervention
period, although their ML and AUC values before are stably
greater than after intervention.
Additionally we compared the increments, i.e., the quotients
of the ML or AUC values after to before presentation between
the two types of pictures. We found a significant picture stimu-
lus effect on the ghrelin increments (Wilks multivariate test of
significance; effect of picture stimulus: F(2,6) = 13.68, sig of F =
0.006), attributed to both indicators ML and AUC (univariate
F-tests, P < 0.05). Figure 2 gives an impression about the incre-
ments of ghrelin after the neutral and food pictures. Neither
for leptin nor for insulin significant picture stimulus effects on
the quotients of the ML or AUC values were found.
As expected, ghrelin decreased after breakfast (mean maxi-
mum ghrelin: neutral picture session at 08:15: 210.0 ± 36.2 pg/
ml; food picture session at 08:30: 234.0 ± 30.9 pg/ml) to the
lowest ghrelin values between 09:30 and 10:30 (mean mini-
mum ghrelin: neutral picture session at 09:30: 166.0 ± 24.8 pg/
ml, food picture session at 09:30: 152.0 ± 21.7 pg/ml). Shortly
after lunch at 12:15, ghrelin reached its maximum in both ses-
sions (neutral picture session: 261.0 ± 44.9 pg/ml, food picture
session: 247.0 ± 35 pg/ml).
Insulin increased to peak after breakfast between 09:15 and
09:45 (mean maximum insulin: neutral picture session at
09:30: 52.9 ± 24.9 pg/ml, food picture session at 09:15: 56.6 ±
49.1 g/ml) and fell steadily before lunchtime to its nadir (mean
minimum insulin: neutral picture session at 12:00: 7.7 ± 4.5
pg/ml, food picture session at 11:45: 8.4 ± 28.1 pg/ml). After
lunch at 12:45, insulin increased again to a postprandial surge
table 1 Mean ± s.e.m. of the area under the curve (auc) of the hormone concentration levels during intervals before (interval 1,
10:00–10:30) and after (interval 2, 10:50–11:20) picture presentation as well as of their increments (quotients of auc after to auc
AUC_neut pict interval 1368.1937.58842.828.148 7.08 1.728
AUC_neut pict interval 2584.0056.94847.018.50810.922.208
F(1,7) = 82.75, P < 0.0001
F(1,7) = 0.039, P = 0.551
F(1,7) = 46.19, P < 0.0001
AUC_food pict interval 1327.9330.97849.479.0486.771.698
AUC_food pict interval 2619.4263.60842.547.67810.502.318
F(1,7) = 72.08, P < 0.0001
F(1,7) = 0.030, P = 0.601
F(1,7) = 29.49, P = 0.001
Quot_neut (interval 2 to
Quot_food (interval 2 to
F(1,7) = 28.43, P = 0.001
F(1,7) = 0.025, P = 0.630
F(1,7) = 0.02, P = 0.887
Ghrelin and leptin but not insulin pointed to significant differences between the two intervals in both presentation modules. But only for ghrelin the increments showed
significant differences between neutral and food pictures (P values < 0.05, by univariate F-tests in MANOVA).
Behavior and Psychology
(neutral picture session: 42.4 ± 24.5 pg/ml, food picture ses-
sion: 36.7 ± 28.1 pg/ml).
Plasma leptin decreased mildly from 08:15 until 09:30. Apart
from that, leptin did not display an interpretable variation in
the course of time.
association between ghrelin, insulin, and leptin
Not any significant correlation was found between the AUC
values of the hormones ghrelin, insulin, and leptin, in each
of the two experimental phases, neither for the neutral nor
for the food picture presentation (Pearson correlation coeffi-
cients. P values n.s., see Table 2). However, when taking into
account the increments of the AUC values after to before pic-
ture presentation, we found a significantly positive association
(r = 0.767, P < 0.05) between insulin and leptin by the pres-
entation of neutral pictures and a significantly negative asso-
ciation (r = −0.736, P < 0.05) between ghrelin and insulin by
the presentation of food pictures. That means that increases
of ghrelin AUC after pictures showing food are related with
decreases of insulin AUC.
Furthermore, we analyzed by contrast tests the single effects,
i.e., interval differences within each picture presentation-
modus and vice versa differences in picture presentation-
modus within each interval and identified that ghrelin levels
during the 30-min interval following the presentation of food
pictures (intervention) increased significantly (more than
18%), compared to the preintervention interval (tests with
contrasts in ANOVA, P < 0.05). In contrast, after presenta-
tion of the neutral pictures, ghrelin levels remained unchanged
(+3%). The difference in the increases between the two condi-
tions was also significant for ghrelin (P < 0.05, Wilcoxon tests
for matched samples).
Insulin and leptin showed no significant differences, neither
between the picture presentation conditions nor between the
Subject 1 did not recognize four of 10 pictures in both sessions.
Subject 5 and 8 made a mistake in one experimental session.
There is no significant correlation between the results of the
recognition test and ghrelin increase.
subjective taste rating
For the set of the 50 food pictures, we found a high correlation
(r = 0.7516, P < 0.0001) between the averaged scoring of the 21
raters and the eight subjects of the study. That indicates a rela-
tive high objectivity in scoring the contents of the pictures.
We did not find a significant correlation between the mean
taste scores and the increase of ghrelin (r = −0.53).
Our major findings are visual presentation of food resulted in
a significant increase of ghrelin (AUC: 30%) compared to the
time interval before the visual presentation. This change dif-
fered significantly from the presentation of neutral pictures,
which did not prompt a significant increase in ghrelin levels.
The course of leptin and insulin concentration was not affected
by the visual presentation. As expected, ghrelin decreased
between 09:30 and 10:30 to its lowest levels and increased at
lunchtime. Insulin raised to peak after meal times. A nega-
tive association between the AUCs of ghrelin and insulin after
presentation of food pictures was found. The taste rating of the
presented food pictures at the end of the experiment was not
correlated to plasma ghrelin levels.
As far as we know, this is the first study demonstrating
an effect on peripheral ghrelin plasma levels, mediated by
an external cue such as visual presentation of hedonic food.
This supports the view that peripheral ghrelin secretion is
regulated by the central nervous system (12). In subjects
fasting for 24 h, ghrelin levels increased and spontaneously
Neutral pictures n=8
Before presentation After presentation
Food pictures n=8
Before presentationAfter presentation
Figure 2 Ghrelin concentration levels of the single individuals before
(interval 1, 10:00–10:30) and after (interval 2, 10:50–11:20) presentation
of neutral and food pictures (n = 8).
Behavior and Psychology
decreased at customary mealtimes (22), which points to a
cephalic mechanism in terms of a circadian rhythm or a cog-
nitive process such as a conditioned reflex. In our study, food
pictures were presented at a time (at 10:30) where food intake
usually is not expected and ghrelin levels reach their lowest
point between breakfast and lunch. Viewing food pictures
at that time resulted in a significant increase in the ghrelin
plasma level compared to the interval before the presentation
of food images, which is a weaker effect than the increase in
ghrelin at mealtimes in our experiment and in previous stud-
ies (10,11). We did not assess the feeling of hunger in our sub-
jects, because we did not want to call their cognitive attention
on appetite-related contents beside the visual presentation.
We suggest that our subjects did not feel very hungry before
the presentation of pictures, because there is evidence that
the temporal course of ghrelin is closely positively correlated
with subjective hunger scores. Low ghrelin plasma levels and
low hunger scores were reported 60–150 min after a meal
(11), the time period where our subjects received the visual
presentation. So we assume that the increase of ghrelin is
related to the visual stimulus and not induced by an internal
cue such as a conditioned time schedule for food intake or a
(preexisting) feeling of hunger. Otherwise, the relative mod-
erate rise of ghrelin after the visual food stimuli indicates that
there are other, maybe additional mechanisms causing the
stronger ghrelin surge at mealtimes.
There is a strong evidence that ghrelin mainly produced in
the stomach induces its orexigenic effects via the vagus nerve
and the brainstem to the hypothalamus (24). In subjects, who
had surgery involving vagotomy, ghrelin did not stimulate food
intake (25). On the other hand, there is an active transport of
ghrelin through the blood–brain interface (26). Perception of
hedonic food by the central nervous system appears to stim-
ulate ghrelin secretion in the stomach. The elevated ghrelin
levels may then induce appetite via the vagus nerve or an active
transport through the blood–brain barrier to the central nerv-
The pictures with food content, which were used as a visual
stimulus for the experiment, were chosen based on rating in
healthy male volunteers, around 2–3 h after food intake, and
illustrated highly preferred savory or sweet food. Subjects par-
ticipating in the ghrelin–visual stimuli experiment rated the
appetizing appearance of the food images shortly after lunch
in a satiated condition. These different states of hunger did
not influence the valuation of the food pictures. Furthermore,
we also did not find a significant correlation between the taste
scores and the increase in ghrelin, though this evidence is sta-
tistically limited by the small number of subjects in the present
Leptin is mainly produced in the adipose tissue, and periph-
eral leptin levels are approximately proportional to body fat
content. A small amount of leptin is secreted by the stomach
(27). Therefore, it could play a role in the short-term regula-
tion of food intake or food digestion/absorption (28). In our
study, we did not find a correlation between leptin and ghre-
lin or leptin and insulin. Leptin plasma levels did not show
any variation after the visual presentation. The mild decrease
in leptin after breakfast is consistent with the observation that
leptin displays a diurnal rhythm with a nadir around 09:00
(10,29). Consistent with previous reports (11,30), the insu-
lin level increased after breakfast within an average of 60 min
and decreased steadily toward their basal value shortly before
lunch. Insulin is thought to have a modulating effect on ghrelin
(31), likewise hyperinsulinemia suppressed peripheral ghrelin
secretion (32) and in turn ghrelin inhibited insulin secretion
(19,33). The negative association between ghrelin and insulin
in our study supports the hypothesis of an insulin suppressing
effect of physiologic doses of ghrelin in humans (34).
table 2 Pearson correlation coefficients of the hormone auc values and their increments (quotients) in the two intervals (before
and after picture presentation)
Association intensity between the hormones
Interval 1Interval 2 Quotients of AUCs (interval 2 to interval 1)
Presentation of neutral pictures
Ghrelin0.1781 0.5853 −0.37860.3054 0.2054 0.5286
P = 0.673
P = 0.127
P = 0.355
P = 0.462
P = 0.626
P = 0.178
Insulin 0.49080.5222 0.7672
P = 0.217
P = 0.184
P = 0.026a
Presentation of food pictures
Ghrelin 0.23190.4312−0.6175 0.3209 −0.73630.5329
P = 0.581
P = 0.286
P = 0.103
P = 0.438
P = 0.037a
P = 0.174
P = 0.595
P = 0.917
P = 0.837
By the presentation of neutral pictures the increments of insulin were significantly positive correlated with the increments of leptin, whereas by the presentation of food
pictures the increments of insulin were significantly negative correlated with those of ghrelin (P values < 0.05).
AUC, area under the curve.
Behavior and Psychology
However, the increase in ghrelin after visual food stimuli
in our study did not modulate insulin secretion. The rising of
insulin levels known to be closely linked with carbohydrate
ingestion (30) was not affected by sight of food per se.
Our finding suggests that an external cue such as sight of
hedonic food elevates ghrelin levels in young, healthy subjects
of normal weight. The exposure to the food stimuli was carried
out at a time point, where subjects were near to the intermedi-
ate state on the continuum hungry to sated. Leptin and insulin
were not affected by the visual presentation. It is possible that
ghrelin is directly controlled by the central nervous system and
that as reported before from other research groups sympathetic
nerves act as a signaling pathway of the gut–brain axis.
In our study, secretion of ghrelin, the strongest orexigenic
hormone known so far, was induced by perception of pictures
with food. This finding supports the hypothesis that environ-
mental factors contribute to eating behavior in modern society,
where the visual presentation of food products is common.
Supplementary material is linked to the online version of the paper at http://
The authors declared no conflict of interest.
© 2012 The Obesity Society
1. Rolls BJ, Rowe EA, Rolls ET. How flavour and appearance affect human
feeding. Proc Nutr Soc 1982;41:109–117.
2. Cornell CE, Rodin J, Weingarten H. Stimulus-induced eating when satiated.
Physiol Behav 1989;45:695–704.
3. Marcelino AS, Adam AS, Couronne T, Köster EP, Sieffermann JM. Internal and
external determinants of eating initiation in humans. Appetite 2001;36:9–14.
4. Valassi E, Scacchi M, Cavagnini F. Neuroendocrine control of food intake.
Nutr Metab Cardiovasc Dis 2008;18:158–168.
5. van der Lely AJ, Tschöp M, Heiman ML, Ghigo E. Biological, physiological,
pathophysiological, and pharmacological aspects of ghrelin. Endocr Rev
6. Tschöp M, Smiley DL, Heiman ML. Ghrelin induces adiposity in rodents.
7. Wren AM, Small CJ, Ward HL et al. The novel hypothalamic peptide ghrelin
stimulates food intake and growth hormone secretion. Endocrinology
8. Wren AM, Seal LJ, Cohen MA et al. Ghrelin enhances appetite and increases
food intake in humans. J Clin Endocrinol Metab 2001;86:5992–5995.
9. Schmid DA, Held K, Ising M et al. Ghrelin stimulates appetite, imagination of
food, GH, ACTH, and cortisol, but does not affect leptin in normal controls.
10. Cummings DE, Purnell JQ, Frayo RS et al. A preprandial rise in plasma
ghrelin levels suggests a role in meal initiation in humans. Diabetes
11. Cummings DE, Frayo RS, Marmonier C, Aubert R, Chapelot D. Plasma
ghrelin levels and hunger scores in humans initiating meals voluntarily
without time- and food-related cues. Am J Physiol Endocrinol Metab
12. Mundinger TO, Cummings DE, Taborsky GJ Jr. Direct stimulation of ghrelin
secretion by sympathetic nerves. Endocrinology 2006;147:2893–2901.
13. Cummings DE, Overduin J. Gastrointestinal regulation of food intake. J Clin
14. Killgore WD, Young AD, Femia LA et al. Cortical and limbic activation during
viewing of high- versus low-calorie foods. Neuroimage 2003;19:1381–1394.
15. Gottfried JA, O’Doherty J, Dolan RJ. Encoding predictive reward value in
human amygdala and orbitofrontal cortex. Science 2003;301:1104–1107.
16. Malik S, McGlone F, Bedrossian D, Dagher A. Ghrelin modulates brain
activity in areas that control appetitive behavior. Cell Metab 2008;7:400–409.
17. Sun Y, Asnicar M, Smith RG. Central and peripheral roles of ghrelin on
glucose homeostasis. Neuroendocrinology 2007;86:215–228.
18. Nogueiras R, Tschöp MH, Zigman JM. Central nervous system regulation
of energy metabolism: ghrelin versus leptin. Ann N Y Acad Sci 2008;
19. Broglio F, Arvat E, Benso A et al. Ghrelin, a natural GH secretagogue
produced by the stomach, induces hyperglycemia and reduces insulin
secretion in humans. J Clin Endocrinol Metab 2001;86:5083–5086.
20. Broglio F, Gottero C, Prodam F et al. Ghrelin secretion is inhibited by glucose
load and insulin-induced hypoglycaemia but unaffected by glucagon and
arginine in humans. Clin Endocrinol (Oxf) 2004;61:503–509.
21. Chapelot D, Aubert R, Marmonier C, Chabert M, Louis-Sylvestre J. An
endocrine and metabolic definition of the intermeal interval in humans:
evidence for a role of leptin on the prandial pattern through fatty acid
disposal. Am J Clin Nutr 2000;72:421–431.
22. Natalucci G, Riedl S, Gleiss A, Zidek T, Frisch H. Spontaneous 24-h ghrelin
secretion pattern in fasting subjects: maintenance of a meal-related pattern.
Eur J Endocrinol 2005;152:845–850.
23. Spiegel K, Tasali E, Leproult R, Scherberg N, Van Cauter E. Twenty-
four-hour profiles of acylated and total ghrelin: relationship with glucose
levels and impact of time of day and sleep. J Clin Endocrinol Metab
24. Murphy KG, Dhillo WS, Bloom SR. Gut peptides in the regulation of food
intake and energy homeostasis. Endocr Rev 2006;27:719–727.
25. le Roux CW, Neary NM, Halsey TJ et al. Ghrelin does not stimulate food
intake in patients with surgical procedures involving vagotomy. J Clin
Endocrinol Metab 2005;90:4521–4524.
26. Banks WA, Tschöp M, Robinson SM, Heiman ML. Extent and direction of
ghrelin transport across the blood-brain barrier is determined by its unique
primary structure. J Pharmacol Exp Ther 2002;302:822–827.
27. Bado A, Levasseur S, Attoub S et al. The stomach is a source of leptin.
28. Klok MD, Jakobsdottir S, Drent ML. The role of leptin and ghrelin in the
regulation of food intake and body weight in humans: a review. Obes Rev
29. Weigle DS, Cummings DE, Newby PD et al. Roles of leptin and ghrelin in
the loss of body weight caused by a low fat, high carbohydrate diet. J Clin
Endocrinol Metab 2003;88:1577–1586.
30. Erdmann J, Töpsch R, Lippl F, Gussmann P, Schusdziarra V. Postprandial
response of plasma ghrelin levels to various test meals in relation to food
intake, plasma insulin, and glucose. J Clin Endocrinol Metab 2004;89:
31. Broglio F, Prodam F, Riganti F, Muccioli G, Ghigo E. Ghrelin: from
somatotrope secretion to new perspectives in the regulation of peripheral
metabolic functions. Front Horm Res 2006;35:102–114.
32. Flanagan DE, Evans ML, Monsod TP et al. The influence of insulin on
circulating ghrelin. Am J Physiol Endocrinol Metab 2003;284:E313–E316.
33. Akamizu T, Takaya K, Irako T et al. Pharmacokinetics, safety, and endocrine
and appetite effects of ghrelin administration in young healthy subjects. Eur J
34. Castañeda TR, Tong J, Datta R, Culler M, Tschöp MH. Ghrelin in the
regulation of body weight and metabolism. Front Neuroendocrinol