Regulation of feeding and anxiety by a-MSH reactive
Maria Hamze Sinnoa, Jean Claude Do Regob, Moı ¨se Coe ¨ffiera,
Christine Bole-Feysota, Philippe Ducrotte ´a, Danie `le Gilbertc, Franc ¸ois Tronc,
Jean Costentinb, Tomas Ho ¨kfeltd, Pierre De ´chelottea, Sergueı ¨O. Fetissova,*
aDigestive System & Nutrition Laboratory (ADEN EA4311), Institute of Biomedical Research, Rouen University & Hospital, IFR23,
76183 Rouen, France
bExperimental Neuropsychopharmacology Laboratory (CNRS FRE 2735), Institute of Biomedical Research, Rouen University &
Hospital, IFR23, 76183 Rouen, France
cImmunopathology Laboratory (Inserm U519), Institute of Biomedical Research, Rouen University & Hospital, IFR23,
76183 Rouen, France
dDepartment of Neuroscience, Karolinska Institutet, 17177 Stockholm, Sweden
Received 13 June 2008; received in revised form 25 August 2008; accepted 25 August 2008
Psychoneuroendocrinology (2009) 34, 140—149
involved in the regulation of motivated behavior, appetite and emotion including stimulation
of satiety and anxiety. Although autoantibodies (autoAbs) reactive with a-MSH have been
identified in human subjects and in rats, it remained unknown if these autoAbs are involved
in the regulation of feeding and anxiety and if their production is related to stress. Here we show
that repeated exposure of rats to anxiolytic mild stress by handling increases the levels and
affinity of a-MSH reactive IgG autoAbs and that these changes are associated with adaptive
feeding and anxiety responses during exposure of rats to a strong stress by food restriction.
Importantly, an increase in affinity of a-MSH reactive autoAbs was associated with changes of
their functional roles from stimulation to inhibition of a-MSH-mediated behavioural responses,
suggesting that these autoAbs can be a carrier or a neutralizing molecule of a-MSH peptide,
purified from blood of rats exposed to repeated mild stress, but not from control rats, are able to
increase acutely food intake, suppress anxiety and modify gene expression of hypothalamic
neuropeptides in naı ¨ve rats. These data provide the first evidence that autoAbs reactive with a-
MSH are involved in the physiological regulation of feeding and mood, supporting a further role of
the immune system in the control of motivated behavior and adaptation to stress.
# 2008 Elsevier Ltd. All rights reserved.
a-Melanocyte-stimulating hormone (a-MSH) is a stress-related neuropeptide
* Corresponding author at: ADEN Laboratory, Faculte ´ de Me ´decine-Pharmacie, 22 Bld Gambetta, Rouen 76183 Cedex 1, France.
Tel.: +33 2 35 14 84 55; fax: +33 2 35 14 82 26.
E-mail address: Serguei.Fetissov@univ-rouen.fr (S.O. Fetissov).
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/psyneuen
0306-4530/$ — see front matter # 2008 Elsevier Ltd. All rights reserved.
senger derived from the precursor proopiomelanocortin
(POMC) expressed mainly in the brain and pituitary (Harris
and Lerner, 1957; Nakanishi et al., 1979). Various stressors,
including psychological stress, stimulates secretion of a-MSH
in the systemic circulation (Wilson and Morgan, 1980) and in
the brain (Liu et al., 2007) resulting in modulation of moti-
vated behavior, including stimulation of satiety (Heisler
et al., 2002; Cone, 2005), anxiety (Datta and King, 1977;
Chaki and Okuyama, 2005) and stress-associated memory
(Wilson and Morgan, 1980). Four melanocortin receptor sub-
types, differentially expressed in organs and tissues, mediate
a-MSH-induced biological effects (Mountjoy et al., 1992;
Schio ¨th, 2001).
Autoantibodies (autoAbs) reactive with a-MSH have been
initially identified in patients with anorexia and bulimia ner-
vosa by their increased binding to a-MSH on the rat brain
sections (Fetissov et al., 2002). Although further studies
revealed that the levels of a-MSH reactive autoAbs correlated
with psychopathological traits in eating disorders (Fetissov
etal., 2005), these autoAbswerealsoreadily detected insera
of healthy individuals and in rats (Fetissov et al., 2005, 2006,
2008) suggesting that a-MSH reactive autoAbs may interfere
with melanocortin signalling in both normal and pathological
conditions. However, the functional role of a-MSH reactive
autoAbs is presently unknown adding to the fact that, in
general, the physiological significance of autoAbs directed
against regulatory peptides remains largely unexplored.
this work, we tested our hypothesis that production of a-MSH
stress and that repeated exposure of animals to stressors will
result in changes of blood levels of a-MSH reactive autoAbs.
Furthermore, to determine if a-MSH autoAbs are involved in
regulation of feeding and anxiety, we studied if stress-related
changes of a-MSH autoAbs in rats will be responsible for
behavioural modifications including feeding and anxiety.
2. Materials and methods
Rat care and experimentation were in accordance with
guidelines established by the National Institutes of Health,
regulations (Official Journal of the European Community L
358, 18/12/1986). Female Wistar rats were purchased from
Elevage Janvier, Le Genest-St-Isle, France and were main-
tained at 24 8C with a 12:12-h light—dark cycle (light period
07:00—19:00 h) in a specialized animal facility. When kept in
standard holding cages (3 rats per cage) rats were fed with
the standard pelleted rodent chow (RM1 diet, SDS, UK), when
kept in metabolism cages (Tecniplast, France) they were fed
with the same RM1 but ground chow (SDS). Drinking water
was always available ad libitum.
2.2. Blood sampling during repeated mild stress
One-month-old rats were randomly divided into four
groups named A, B, C, D (n = 6 in each group) and were
kept in holding cages with food available ad libitum (for
experimental design see Supplementary Fig. 1). Body
weight was monitored weekly. After one week of acclima-
tization, rats of the group A were anaesthetized by an
intraperitoneal (IP) injection (25 mg/kg body weight) of
Brietal (Lilly, Netherlands). During the brief anaesthesia, a
200 ml blood sample was taken by retroorbital puncture
(Bleed-1) and rats were returned to their cages. Two weeks
later, blood sampling was performed in the A and B groups
(Bleed-2), and in two more weeks in groups A, B and C
(Bleed-3). During this schedule, the group A underwent
three blood sampling, which can be considered as a
repeated mild stress. The two-week interval between
sampling was necessary to observe maximal autoAbs
response. The day after the Bleed-3, rats of the groups
A, B, C and D were placed into the individual metabolic
cages and were food-restricted for six days with food
available 1 h/day. Body weight at the start of food restric-
tion schedule was measured for four groups (mean
g ? S.D.; 187 ? 5, 190 ? 13, 191 ? 21, 189 ? 21, respec-
tively, ANOVA p = 0.98). At the last day of food restriction,
rats passed anxiety test in the elevated plus-maze,
then they were anaesthetized by sodium pentobarbital
(2 mg/kg body weight, IP) and 2 ml blood samples were
taken from the right atrium (Bleed-4). Blood samples were
also taken from ad libitum (Ad lib) fed rats of the same
age (n = 6).
2.3. a-MSH autoantibody assay
Serum levels of IgG or IgM autoAbs reacting with a-MSH
were measured using enzyme-linked immunosorbent assay
(ELISA) technique. The a-MSH peptide (Bachem AG, Buben-
dorf, Switzerland) was coated on Maxisorp plates (Nunc,
Rochester, NY) using 100 ml and a concentration of 2 mg/ml
in 100 mM NaHCO3buffer, pH 9.6 for 24 h at 4 8C. Plates
were washed (5 min ? 3) in phosphate-buffered saline (PBS)
with 0.05% Tween 200, pH 7.4, and then incubated over-
night at 4 8C with 100 ml of rat sera diluted 1:200 in PBS to
determine free autoAbs levels or diluted 1:200 in dissocia-
tive 3 M NaCl, 1.5 M glycine buffer, pH 8.9 to determine
total autoAbs levels. The optimal dilutions of sera (1:200)
were determined by dilution curves (1:50, 1:100, 1:200,
1:400 and 1:800). The plates were washed (3?) and for the
detection of a-MSH IgG autoAbs incubated with 100 ml of
alkaline phosphatase-conjugated goat anti rat IgG (1:2000)
(Sigma, St. Louis, MO) for 3 h at room temperature (RT). For
the detection of a-MSH IgM autoAbs, monoclonal anti rat
IgM, clone RTM-32 (Sigma) was diluted in PBS (1:2000) and
incubated for 3 h at RT. Then, alkaline phosphatase-con-
jugated rabbit anti mouse IgG (Jackson ImmunoResearch
Europe Ltd., Cambridgeshire, UK) diluted in PBS 1:500 was
added for 2 h at RT. Following washing (3?), 100 ml of p-
nitrophenyl phosphate solution (Sigma) was added as alka-
line phosphatase substrate. After 40 min of incubation at
RT, the reaction was stopped by adding 3N NaOH. The
optical density (OD) was determined at 405 nm using a
microplate reader. Blank OD values resulting from the
reading of plates without addition of rat sera were sub-
tracted from the sample OD values. Each determination was
done in duplicate. The variation between duplicate values
was less than 5%.
MSH autoantibodies in feeding and anxiety 141
2.4. a-MSH peptide assay
Serum level of a-MSH peptide was measured in the Bleed-4
samples of all rats using a-MSH EIA kit (EK-043-01) purchased
from Phoenix Pharmaceuticals, Inc, Belmont, CA. The mini-
mum detectable concentration of this assay was 0.13 ng/ml
with the linear range till 1.62 ng/ml as indicated by the
2.5. Anxiety test
The elevated plus-maze test was performed according to
Pellow et al. (1985). The apparatus for rats consisted of a
wooden Greek cross, painted black and placed 80 cm above
the floor in a dimly illuminated room. The four arms (dimen-
sion of each arm: L: 110 cm, W: 10 cm) were interconnected
by a central platform (10 cm ? 10 cm). Two opposite arms
were surrounded by walls (H: 31 cm; closed arms) while the
two others were devoid of enclosing walls (open arms). The
light intensity at the level of the elevated plus-maze was
100 lux. Each rat was placed at the center of the maze, head
facing a closed arm. The time spent in open and closed arms
and in the central area during a 5-min period was recorded
using an automated image analysis system (Videotrack MV 45
system, Viewpoint, Lyon, France). The rat was considered to
be in the open or in the closed arms or in the central area
when the gravity center of its image (seen by the camera, set
around 1 m above the maze) was detected in each of these
2.6. Autoantibody purification and affinity assay
Total IgG was purified from sera of rats from the group A and
D using protein G agarose (Sigma) according to manufac-
turer instructions. Affinity of rat IgG autoAbs for a-MSH was
determined by a biospecific interaction analysis (BIA) based
on the surface plasmon resonance phenomenon in the
BIAcore 1000 instrument (BIAcore, GE Healthcare, Piscat-
away, NJ). a-MSH peptide (Bachem) was diluted 1 mg/ml in
10 mM sodium acetate buffer and was covalently coupled on
the sensor chip CM5, (BIAcore) using the amine coupling kit
(BIAcore). 25 ml of total IgG from individual samples of
group A or Group D rats were diluted to the same concen-
tration and were injected into the flow conduit of BIAcore
1000 instrument. The values of resonance units (RU)
reflecting the affinity of autoAbs were recorded at the
plateau of the dissociation curve of the sensorgram (at
equilibrium) in 5 min after injection at 25 8C with flow
speed 5 ml/min.
Following the individual affinity assays, the sera purified
from rats of the group A and D was combined into two pools A
and D, respectively. a-MSH IgG autoAbs from pools A and D
were purified using affinity chromatography with a-MSH
peptide (Bachem) coupled to the pre-activated beads (Ultra-
Link, Pierce, Rockford, IL) according to manufacturer’s
instructions. The purified a-MSH IgG autoAbs were lyophi-
lized and diluted in artificial cerebrospinal fluid (aCSF) to the
concentration of 5 mg/ml in both pools. Presence of a-MSH
autoAbs in the eluted IgG fraction was confirmed by ELISA as
described above. BIAcore assay for a-MSH affinity of the Pool-
A and Pool-D purified a-MSH autoAbs was performed again
using the same parameters as described above.
2.7. Passive transfer of a-MSH autoAbs
Two-month-old female Wistar rats (n = 24) were placed into
individual metabolic cages with food and water available ad
libitum (Supplementary Fig. 2). After one week of acclima-
tization, rats were anaesthetized (1 ml/kg body weight, IP)
with a mixture (3:1 vol.) of Ketamine (10%) and Xylazine (2%)
and stereotactically implanted (900 Stereotaxic apparatus,
David Kopf Instruments, Tujunga, CA) with stainless steel
cannulas for acute injections (C311 GA, diameter external
0.9 mm, internal 0.58 mm, Plastics One, Roanoke, VA) aimed
to the right ventromedial hypothalamus (Stereotactic coor-
dinates: the incisor bar ?3.3 mm, Bregma ?2.4 mm, lateral
from the saggital sinus 0.5 mm, ventral from the skull
8.5 mm) (Swanson, 1998). The cannulas were permanently
fixed to the skulls with dental cement and anchoring screws
and were closed with dummy cannulas (C311 DC, Plastics
One). After seven days, when all rats recovered from the
operation as manifested by their return to the pre-operation
level of food intake for at least three consecutive days, rats
were randomly divided into three groups (n = 8): Pool-A,
Pool-D and Control and were food-restricted for five days
libitum. Food intake was measured for 1 h and for the
following 7 h after food provision. Starting from the second
day of food restriction and 1 h prior to food provision Pool-A,
Pool-D and Control rats received injections via the brain
cannula of 1 ml of a-MSH IgG purified from the group A, from
the group D or aCSF, respectively, using 5 ml syringes without
dead volume (Hamilton, Reno, NV). The needle of the syringe
was equipped with a plastic stopper which ensured that the
tip of the needle was located at the end of the cannula.
At the last day of food restriction, rats received another
hypothalamic injection of 1 ml of a-MSH IgG autoAbs or aCSF
and 15 min later were placed into the plus-maze for anxiety
test as described above. Immediately after the plus-maze
test, to measure mRNA expression of hypothalamic neuropep-
tides, all rats were anaesthetized by sodium pentobarbital
(2 mg/kg body weight, IP) and were killed by decapitation.
Brains were harvested, entire hypothalami were dissected,
placed in 1 ml of guanidinium thiocyanate solution and frozen
in liquid nitrogen until further assay.
2.8. Hypothalamic neuropeptide mRNA assay
Expression levels of mRNA of several neuropeptides involved
in appetite control were measured by quantitative polymer-
ase chain reaction (qPCR) on whole hypothalamic homoge-
nates. Briefly, mRNA was isolated according to the protocol
adapted from Chomczynski and Sacchi (1987). Individual
hypothalamic tissue samples were homogenised in 1 ml of
guanidinium thiocyanate solution, then 1 ml of phenol,
250 ml of chloroform and 100 ml of Na acetate were added.
The homogenates were centrifuged at 12,000 ? g for 20 min
RNA was precipitated with isopropanol overnight at ?20 8C.
The pellets were dissolved with 75% ethanol and dried in a
Speed-Vac Concentrator (Savant Instruments, Farmingdale,
NY), then dissolved in RNAse free water. RNA quality was
confirmed by 0.8% agarose gel electrophoresis stained by
ethidium bromide. RNA concentration was measured on
142 M.H. Sinno et al.
Reverse transcription was used to synthesize cDNA, with
reverse transcriptase and random hexamers as primers
following the manufacturer instructions (Invitrogen, Carl-
sab, CA). The quality of the cDNA was confirmed by PCR
using a primer pair for glyceraldehyde 3-phosphate dehy-
drogenase (GAPDH) as an internal control. Relative levels of
mRNA were determined by the qPCR using a LightCycler
instrument (Roche, Basel, Switzerland). qPCR was per-
formed using absolute SYBR Green Capillary Mix (Abgene,
Epsom, UK) in a 20 ml of final reaction volume. Primer pairs
(Supplementary Table 1) were designed using Primer Pre-
mier 5.0 program (Biosoft Int, Palo Alto, CA). Prior to the
qPCR, the correct size of PCR product for each primer pair
was verified by classical PCR. All PCR products matched
with their designed sizes. All samples were measured in
duplicates. Equal amounts of cDNA were added to each
reaction (equivalent to 2 mg of total RNA). The following
PCR conditions were used: initial denaturation at 95 8C for
15 min, amplification over 40 cycles with: denaturation at
95 8C for 15 s, annealing at Tm (Table 1S) for 20 s, and
extension at 72 8C for 20 s. A 5 fold serial dilution of each
gene was assayed from DNA purified on NucleoSpin extract II
columns (Machery-Nagel, Du ¨ren, Germany) for DNA extrac-
tions to construct a standard curve.
Values for each neuropeptide mRNA measured in dupli-
cates were normalised and their levels relative to the level of
the GAPDH mRNA used as internal control from the same
sample was calculated. The level of the GAPDH mRNA did not
differ between the groups (ANOVA p = 0.96).
2.9. Statistical analysis
Data were analyzed in the GraphPad in Stat program (Graph-
Pad Software Inc., San Diego, CA). Data from autoAbs detec-
tion studies were analyzed by repeated measurements
ANOVA parametric or non-parametric (Friedman) tests as
well as by paired t-test between two bleeds in Group C.
Differences in food intake and body weight changes were
analyzed by two-way repeated measurement ANOVAwith the
Bonferroni post hoc test. Other analysis included parametric
or non-parametric (Kruskal—Wallis) ANOVA tests with post
hoc Tukey—Kramer or Dunn’s tests, respectively according to
the normality test. Additionally, individual groups were com-
pared by the Student’s t-test or the Mann—Whitney test
according to the normality test. In all cases, p < 0.05 was
considered to be statistically significant.
3.1. Effect of stress in production of a-MSH
To investigate the role of stress in production of a-MSH
reactive autoAbs, we studied rats exposed to decrement
amount of repeated mild stress by handling and subsequently
rats were randomly divided into four groups named A, B, C, D
(n = 6 in each group). Group A rats were handled for 30 s,
anaesthetized and a 200 ml blood sample was taken by retro-
orbital puncture (Bleed-1). Two weeks later, handling and
blood sampling were similarly performed in A and B groups
(Bleed-2), and again two weeks later in groups A, B, and C
(Bleed-3). The two-week interval between sampling was
chosen to obtain a maximal autoAbs response, since IgG
secretion normally peaks around two weeks after stimulation
by soluble protein antigens. The day after Bleed-3, all rats
were food-restricted for six days with food available 1 h/day,
a situation representing a strong stress. In this experimental
design, group A was exposed the most to a repeated mild
stress, while group D was not exposed at all to the mild stress
before food restriction. Body weight at the start of the food
restriction schedule did not differ among the four groups
We found that immunoglobulin (Ig)G and IgM autoAbs
reactive against a-MSH were present in the sera of all rats
and in all bleeds (Fig. 1B—E). By diluting sera in either a
dissociative buffer or in phosphate-buffered saline, we mea-
sured the levels of total vs. free (unbound) a-MSH autoAbs,
respectively. The presence of higher levels (around 3-fold for
IgG and 2-fold for IgM) of total vs. free a-MSH autoAbs
repeated mild stress applied before Bleeds 1, 2 and 3 and strong
stress by food restriction before final Bleed-4 (for details see
text). (B—E) Serum levels (optical density (OD), mean ? S.E.) of
the total and free a-MSH IgG and IgM autoAbs measured in Bleeds
1—4. Repeated bleeds were characterized by increasing levels of
IgG(BandC)and IgM(Dand E)a-MSHautoAbs. Levels oftotalbut
not free a-MSH IgG autoAbs were higher in group A vs. other
groups in Bleed-4 (B vs. C). Repeated ANOVA, Tukey—Kramer test
ANOVA*p < 0.05.
(A). Body weight change and experimental design of
#p < 0.05,
##p < 0.01, Paired t-test
**p < 0.01,
MSH autoantibodies in feeding and anxiety143
suggests that the majority of the autoAbs existed in the form
of immune complexes (Fig. 1 cf. B vs. C and D vs. E).
The levels of both total and free a-MSH autoAbs in groups
A and B were higher in later vs. earlier bleeds, thereby
showing the positive dynamics of autoAbs secretion asso-
ciated with exposure to mild stress. Moreover, in the final
sampling, Bleed-4, the levels of total but not free a-MSH IgG
autoAbs in group A were significantly elevated vs. the other
groups, indicating an increased immune complex formation
in group A rats (Fig. 1B). Further, increased levels of IgM a-
MSH autoAbs in Bleed-4 vs. Bleed-3 were found in all groups
(Fig. 1D), indicating that secretion of a-MSH induced by
strong stress may also stimulate the production of a-MSH
IgM autoAbs, similar to the normal fast IgM response occur-
ring following an antigenic stimulation.
Serum levels of free a-MSH peptide measured in final
Bleed-4 were significantly lower in group A vs. other groups
(Fig. 2A). Accordingly, positive correlations were found
between levels of a-MSH peptide and free a-MSH IgG autoAbs
in groups B—D (Person’s r = 0.6, p = 0.02) but not in group A,
(Fig. 2B), suggesting that a-MSH autoAbs in groups B—D were
not neutralizing but rather involved in transportation of a-
MSH peptide detected by immunoassay.
3.2. Effects of stress on feeding with relation to
During food restriction, rats from group A lost less body
weight vs. group D (Fig. 3A), most likely a result of differ-
ences in food intake. In fact, although all rats progressively
increased their 1 h food intake during the six-days restriction
period, group A rats displayed higher food intake already
from the onset of restriction (Fig. 3B), resulting in a larger
amount of total food consumed (Fig. 3C). Significant correla-
tions between the levels of both IgG and IgM total a-MSH
autoAbs and food intake were found in rats of groups A and D
(Fig. 3D and E), being positive and negative, respectively,
autoAbs in group A, but stimulated in group D.
3.3. Effects of stress on anxiety with relation to
We found that rats in group A and B displayed less anxious
behavior vs. group D as mainly shown by less time spent
within the closed arms of the plus-maze (Fig. 4A). Signifi-
cant correlations (Pearson’s r = 0.9, p < 0.05) between the
levels of a-MSH IgG autoAbs and time spent in the plus-maze
were found, being positive for open and for closed arms in
groups B and D, respectively (Fig. 4C). These results show
that a-MSH autoAbs could be anxiolytic in less anxious, but
anxiogenic in more anxious rats, suggesting their melano-
cortin antagonist-like role in group B but agonist-like role in
3.4. Effects of stress on affinity of a-MSH
Taken together, both food intake and plus-maze data
showed that the levels of a-MSH autoAbs correlated with
feeding and anxiety. However, the opposite correlations
found in the groups of rats with significant differences in
behavior suggest that the a-MSH autoAbs in these groups
may not have identical properties with relation to a-MSH
To address this question, we measured affinity of IgG
autoAbs for a-MSH using a BIA assay. Total IgG were purified
from the Bleed-4 sera of rats from groups A and D. The total,
purified IgG was combined into two pools, A and D, and a-
MSH IgG autoAbs were further purified using affinity chro-
matography. The presence of a-MSH autoAbs in the eluted
IgG fraction was confirmed by ELISA. We found that free
(before affinity purification) IgG autoAbs in group A had
lower affinity for a-MSH than in group D (resonance units,
mean ? S.E., 289 ? 63 vs. 495 ? 20, respectively, Student’s
t-test p < 0.01). However, the affinity assay performed in
total affinity or avidity values for a-MSH in group A were
around two times higher than in group D.
3.5. Feeding and anxiety in a model of passive
transfer of a-MSH autoAbs
To determine if a-MSH autoAbs can cause changes in food
intake and anxiety, depending on their affinity, we devel-
oped a model of passive transfer of IgG autoAbs, derived
tide in female Wistar rats measured in the Bleed-4. One group
(Ad lib) of female Wistar rats (n = 6) of the same age, was fed ad
libitum before the blood sampling. K—W test p = 0.04, Dunn’s
post hoc test Group A vs. Group C p < 0.05. Student’s t-test
*p < 0.05,**p < 0.01. Of note, the levels of a-MSH autoAbs in
group D was not significantly different from that of rats of the
same age fed ad libitum (data not shown), indicating that our
schedule of food restriction per se was not sufficient to change
the levels of a-MSH autoAbs. (B) Linear correlations (Pearson’s r)
between levels of a-MSH peptide and a-MSH IgG or IgM free
autoAbs. Significant Pearson’s r,*p < 0.05.
(A) Serum levels (mean ng/ml ? S.E.) of a-MSH pep-
144 M.H. Sinno et al.
from the A and D groups, respectively. The affinity purified
a-MSH autoAbs or aCSF were injected into the ventromedial
hypothalamus of three groups of naı ¨ve female Wistar rats
starting from the second day of a five-day food-restriction
schedule. At the first day of food restriction, there were no
differences in food intake among the groups (Fig. 5A—F).
During the injection period, food intake in Pool-A rats was
higher during the first hour after food provision (Fig. 5A and
B), but lower during the following 7 h (Fig. 5C and D) vs.
Controls. However, the ‘bulimic’ response in the Pool-A rats
during the first hour was not sufficient to increase the total
food intake to the same level as seen in Control rats (Fig. 5E
and F), resulting in a significant reduction in body weight
gain (Fig. 5G and H). Feeding response in Pool-D rats was not
significantly different from the Control group, and the 7 h
food intake in this group was higher than in Pool-A rats
(Fig. 5D). Twenty four hour water intake was also signifi-
cantly lower in Pool-A rats vs. Control or Pool-D rats follow-
ing intrahypothalamic injections (Supplementary Fig. 3).
Thus, while Control and Pool-D rats adapted to the food
restriction schedule by an everyday gradual increase of
food intake distributed throughout the day, Pool-A rats
displayed a bulimic response only shortly after food provi-
sion which can be attributed to the blocking effect of
natural a-MSH-induced satiety by the injected high affinity
a-MSH autoAbs. However, this temporally blocked satiety
appears to deregulate the total energy balance in Pool-A
rats, since they were not able to gain body weight at the
same rate as Controls.
At the last day of food restriction, the rats received a final
injection of a-MSH IgG autoAbs or aCSF, and assessed in the
plus-maze. Pool-A rats displayed less anxious behavior vs.
Controls, showing reduced number of entries and shorter
distance in the closed arms, while the behavior of Pool-D rats
did not change significantly from Controls (Fig. 5I and J and
Supplementary Fig. 4).
restriction. All rats increased food intake during restriction but Group A rats had higher food intake than Group D rats already during
first 3 days, resulting in (C) more total food consumed during restriction. (D) Linear correlations (Pearson’s r, mean for six days ? S.E.)
between food intake during restriction and levels of a-MSH IgG or (E) IgM autoAbs measured in the Bleed-4. Significant correlations
were found in the Group A at the Day 2 of food restriction (Pearson’s r = 1.0 p < 0.01), in the Group C at the Day 4 of food restriction
(Pearson’s r = ?0.86 p < 0.05) and in the Group D at the Day 2 of food restriction (Pearson’s r = ?0.84, p < 0.05). K—W test between
means of correlations, p < 0.001, Dunn’s post hoc test*p < 0.05,**p < 0.01.
(A) Body weight loss (g, mean ? S.D.) during six days of food restriction. (B) Food intake (g, mean ? S.D.) during food
Group A rats were less anxious vs. Group D rats by spending less
time in the closed arms but more time in open arms. Group B
showed intermediate results. Student’s t-test
**p < 0.01.(C)Linearcorrelations(Pearson’sr)betweenduration
spent in the plus-maze open or closed arms and levels of free a-
MSH IgG autoAbs measured in the Bleed-4 in four groups of
female Wistar rats. Significant Pearson’s r,*p < 0.05.
(A and B) Anxiety test in the plus-maze showed that
*,#p < 0.05,
MSH autoantibodies in feeding and anxiety145
3.6. Neuropeptide mRNA expression in a model
of passive transfer of a-MSH autoAbs
Increased levels of orexigenic neuropeptide Y (NPY) mRNA
and decreased levels of POMC mRNAwere found in the Pool-A
vs. control rats, most likely reflecting the negative energy
balance present in the Pool-A rats (Fig. 6). However, the
transcript levels of two orexigenic neuropeptides, including
agouti-related protein (AgRP) and melanin-concentrating
hormone (MCH) were significantly lower in Pool-A rats, sug-
gesting that these changes may underlie the deficient total
food intake observed following the administration of a-MSH
autoAbs. No significant differences in orexin (hypocretin) or
corticotrophin-releasing hormone (CRH) mRNA levels were
found, indicating that the state of wakefulness or the stress
axis, respectively, were not involved in the differential
behavioural responses of the Pool-A rats.
Our data show that rats naturally display circulating a-MSH
autoAbs which exist in two pools, characterized by low and
IgG autoAbs on levels of mRNA expression of appetite-regulating
neuropeptides in the hypothalamus of Pool-A, Pool-D and Control
rats on the last day of food restriction. The levels of orexigenic
NPY (A) were increased and the levels of POMC (B) a precursor of
anorexigenic a-MSH were decreased witnessing the negative
energy balance in Pool-A rats vs. Controls. The levels of two
orexigenic neuropeptides, agouti-related protein (AgRP, C) and
melanin-concentrating hormone (MCH, D) were lower in Pool-A
vs. control rats, while levels of orexin (E) and corticotrophin-
releasing hormone (CRH, F) did not change significantly. (values,
mean ? S.E., Student’s t-test*p < 0.05,**p < 0.01).
Effects of administration of affinity purified a-MSH
MSH IgG autoAbs affinity purified from Group A rats (Pool-A) or
from Group D rats (Pool-D) and of aSCF (Control) into the ven-
tromedial hypothalamus (hpt) of naı ¨ve female Wistar rats on food
groups before autoAbs injection, Day 1. (B) During autoAbs injec-
rats, but (D) lower during following seven hours vs. Controls,
resulting in (F) insufficient total food intake in Pool-A rats to gain
body weight (H) at the same rate as Controls. ANOVA,*p < 0.05,
**p < 0.01;pairedStudent’st-tests#p < 0.05,##p < 0.01.(IandJ)
Pool-A rats displayed less anxious behaviour vs. Controls by show-
ing reduced number of entries and shorter distance inside the
closed arms of the plus-maze. Student’s t-test*p < 0.05.
Effects of administration (starting from Day 2) of a-
146 M.H. Sinno et al.
high affinity, respectively, with regard to their ability to form
immune complexes with a-MSH. The effect of chronic mild
stress can be anxiolytic or anxiogenic depending on the
intensity and type of stressor (Zelena et al., 1999; Sergeyev
et al., 2005). In this work we used handling as a model of
in rats (Meerlo et al., 1999; Maslova et al., 2002) a finding
reproduced in our study. We found that repeated exposure of
animals to mild psychological stress is accompanied by
increased levels of complex-forming, high affinity a-MSH
IgG autoAbs and by reduction in affinity of the free fraction
of these autoAbs. Importantly, these changes in level and
affinity were associated with behavioural responses charac-
teristic for reduced melanocortin system-mediated activity,
i.e. increased food intake and lower anxiety. These data
suggest that affinity maturation of a-MSH reactive autoAbs,
triggered by a-MSH secretion during repeated mild stress can
participate in development of adaptive stress-associated
behavioural responses. In fact, passive transfer of higher
affinity a-MSH autoAbs from rats displaying increased food
intake and lower anxiety rendered the recipient rats bulimic
and less anxious as well, providing evidence that autoAbs can
be causative for these behavioural responses attributed to
more efficient blockage of the physiological a-MSH-mediated
satiety and anxiety. In contrast, transfer of lower affinity a-
MSH autoAbs was not able to trigger significant changes in
behavior, probably because their transporting/blocking
action was not sufficient to modulate the intrinsic activity
of the melanocortin system in recipient rats. However, our
data showing that presence of lower affinity a-MSH autoAbs
in rats which were not exposed to repeated mild stress were
associated with their agonist-like activity on the melanocor-
tin system suggest that low affinity autoAbs might play a role
of a-MSH carrier. In fact, a carrier role of autoAbs directed
against some messenger peptide molecules including cyto-
kines and nerve growth factor has previously been proposed
(Dicou and Nerriere, 1997; Bendtzen et al., 1998). We spec-
ulate that a balance between blocking and carrying roles of
a-MSH autoAbs may maintain normal levels of transmission in
the melanocortin system, allowing adaptation to stress via
affinity maturation of the free fraction of autoAbs as sum-
marized in Fig. 7.
Furthermore, strong stress by food deprivation was
associated with increased secretion of IgM a-MSH autoAbs
in all groups of rats. Since IgM class of immunoglobulins
exist in pentamers and is typically characterized by low
affinity towards their antigens, our data suggest that a-
MSH IgM autoAbs may represent the main a-MSH carrier
molecule and therefore will be responsible for amplifica-
tion of a-MSH-mediated stress responses. Interestingly,
increased levels of IgM but not IgG a-MSH autoAbs were
found in patients with anorexia nervosa (Fetissov et al.,
The results of the passive transfer a-MSH IgG autoAbs as
well as significant correlations between serum levels of a-
MSH autoAbs and behavioural responses suggest that circu-
lating a-MSH autoAbs may participate in the brain control of
feeding and anxiety. In fact, in physiological conditions, a-
MSH autoAbs can influence feeding and anxiety by interfering
with the brain melanocortin circuitries via circumventricular
organs including the hypothalamic arcuate nucleus (Cowley
et al., 2001; Sun et al., 2007) and the area postrema both of
which are brain gates for appetite-regulating hormones
(Chaudhri et al., 2006). This suggestion is corroborated by
several studies showing that peripheral neuropeptides parti-
cipate in regulation of body weight and mood (Belzung et al.,
2006; Kuo et al., 2007) and administration of melanocortin
receptor agonists reduces food intake (Marsh et al., 1999;
Trivedi et al., 2003) while immunization-induced increase of
systemic concentration of antibodies against ghrelin, an
appetite-regulating hormone, had significant effect on body
weight gain in rats (Zorrilla et al., 2006). Moreover, immu-
nization of rats with protein fragments of melanocortin MC4
receptor resulted in increased food intake and body weight
(Peter et al., 2007).
Furthermore, although the main aim of our model of
passive transfer of a-MSH autoAbs into the hypothalamus
was to determine the causality relation between changes of
affinity of these autoAbs and motivated behavior, it shows
that access of a-MSH autoAbs to the brain areas normally
protected by the blood—brain barrier can alter the melano-
cortin brainsystem beyond thephysiological regulation (Kishi
and Elmquist, 2005), resulting in alteration of feeding and
anxiety. In fact, in some pathological conditions, circulating
autoAbs can reach brain areas normally protected by the
blood—brain barrier resulting in neurological impairments
which may underlie a number of neurological diseases (Lang
et al., 2003; Steinman, 2004; Dantzer and Kelley, 2008).
melanocortin signaling and adaptation to stress. The peptide a-
MSH is secreted into in the systemic circulation and will be
rapidly (within minutes) degraded by serum proteases unless
protected by a carrier protein, such as autoAbs which display low
(left) or high (right) affinity for a-MSH. The former transport a-
MSH and can easily ‘release’ the peptide to act on melanocortin
receptors. In contrast, high affinity autoAbs retain a-MSH in
bound form, preventing signaling through the receptor and
resulting in degradation in the liver. Low affinity a-MSH autoAbs,
as any immunoglobulin molecule, can undergo affinity matura-
tion triggered by repeated stimulation by the specific antigen.
During stress the elevated secretion of a-MSH will result in
increased production of high affinity autoAbs with neutralizing
function. They will attenuate melanocortin signaling during
exposures to stress, resulting in reduced anxiety and increased
feeding, thereby serving as a molecular link in stress adaptation.
Summary on the involvement of a-MSH autoAbs in
MSH autoantibodies in feeding and anxiety147
We found that administration of high affinity a-MSH
autoAbs in the hypothalamus produced acutely bulimic
response which was followed by a reduced food and water
intake relative to control rats resulting in a lower body
weight gain, possibly as a cumulative effect of lower food
intake. This phenomenon can be related to the complex
hypothalamic mechanism of coordinated regulation of
meal size and meal frequency (Guijarro et al., 2006) which
was deregulated by autoAbs blocking the melanocortin
signalling, the final common pathway of satiety (Cone,
2005). Therefore the initial orexigenic response appears
as an immediate consequence of autoAbs blocking satiety
effect of a-MSH while, at the long term, the total reduc-
tion of appetite and body weight can be related to changes
of gene expression of appetite regulating neuropeptides.
Indeed, we found that repeated delivery of a-MSH autoAbs
into the brain resulted in a decrease of gene expression of
AgRP and MCH, two orexigenic neuropeptides present in
neurons which normally receive a-MSH innervation (Elias
et al., 1998). Although NPY and AgRP are expressed by the
same neurons in the hypothalamic arcuate nucleus (Hahn
et al., 1998), our results indicate a differential regulation
of their expression, supporting earlier data (Kas et al.,
2005), and suggesting distinct roles of these neuropeptides
in stress-related behavior. Hence, it is possible that access
of high affinity autoAbs in the brain regions normally
protected by the blood—brain barrier may potentially
trigger the development of eating disorders. In fact, levels
of a-MSH reactive autoAbs were associated with psycho-
logical traits in patients with anorexia and bulimia (Fetis-
sov et al., 2005).
In conclusion, these data provide the evidence that auto-
Abs reactive with a-MSH, a stress-related neuropeptide, are
involved in regulation of feeding and anxiety and that stress-
induced increase in high affinity a-MSH autoAbs may repre-
sent a mechanisms of adaptation to stress via change of
functional role of these autoAbs from carrier to neutralizing
molecules resulting in increased inhibition of a-MSH-
mediated satiety and anxiety. Considering the growing evi-
dence of cytokines involvement in brain functions (Dantzer
et al., 2008) our results point to a further role of the immune
system in the control of motivated behavior.
Role of funding source
This project was supported by the French Ministry of
Research and High Education; by GPDN Association, Rouen,
France; by Torsten and Ragnar So ¨derberg’s Foundation, Swe-
denand byNARSAD:The mentalHealth ResearchAssociation,
Conflict of interest
There is no conflict of interest associated with this work.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.psyneuen.
Belzung, C., Yalcin, I., Griebel, G., Surget, A., Leman, S., 2006.
Neuropeptides in psychiatric diseases: an overview with a parti-
cular focus on depression and anxiety disorders. CNS Neurol.
Disord. Drug Targets 5, 135—145.
autoantibodies to cytokines. Immunol. Today 19, 209—211.
Chaki, S., Okuyama, S., 2005. Involvement of melanocortin-4 recep-
tor in anxiety and depression. Peptides 26, 1952—1964.
Chaudhri, O., Small, C., Bloom, S., 2006. Gastrointestinal hormones
regulating appetite. Philos. Trans. R. Soc. B: Biol. Sci. 361, 1187—
Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isola-
tion by acid guanidinium thiocyanate-phenol-chloroform extrac-
tion. Anal. Biochem. 162, 156—159.
Cone, R.D., 2005. Anatomy and regulation of the central melano-
cortin system. Nat. Neurosci. 8, 571—578.
Cowley, M.A., Smart, J.L., Rubinstein, M., Cerdan, M.G., Diano, S.,
Horvath, T.L., Cone, R.D., Low, M.J., 2001. Leptin activates
anorexigenic POMC neurons through a neural network in the
arcuate nucleus. Nature 411, 480—484.
Dantzer, R., Kelley, K.W., 2008. Brief Commentary. Autistic children:
a neuroimmune perspective. Brain Behav. Immunity22, 804—805.
Dantzer, R., O’Connor, J.C., Freund, G.G., Johnson, R.W., Kelley,
K.W., 2008. From inflammation to sickness and depression: when
the immune system subjugates the brain. Nat. Rev. Neurosci. 9,
Datta, P.C., King, M.G., 1977. Effects of melanocyte-stimulating
hormone (MSH) and melatonin on passive avoidance and on an
emotional response. Pharmacol. Biochem. Behav. 6, 449—452.
Dicou, E., Nerriere, V., 1997. Evidence that natural autoantibodies
NGF. J. Neuroimmunol. 75, 200—203.
Elias, C.F., Saper, C.B., Maratos-Flier, E., Tritos, N.A., Lee, C., Kelly,
J., Tatro, J.B., Hoffman, G.E., Ollmann, M.M., Barsh, G.S.,
Sakurai, T., Yanagisawa, M., Elmquist, J.K., 1998. Chemically
defined projections linking the mediobasal hypothalamus and the
lateral hypothalamic area. J. Comp. Neurol. 402, 442—459.
Fetissov, S.O., Hallman, J., Nilsson, I., Lefvert, A.K., Oreland, L.,
Ho ¨kfelt, T., 2006. Aggressive behavior linked to corticotropin-
reactive autoantibodies. Biol. Psychiatry 60, 799—802.
Fetissov, S.O., Hallman, J., Oreland, L., af Klinteberg, B., Grenba ¨ck,
E., Hulting, A.L., Ho ¨kfelt, T., 2002. Autoantibodies against a-
MSH, ACTH, and LHRH in anorexia and bulimia nervosa patients.
Proc. Natl. Acad. Sci. U. S. A. 99, 17155—17160.
Fetissov, S.O., Hamze Sinno, M., Coe ¨ffier, M., Bole-Feysot, C.,
Ducrotte ´, P., Ho ¨kfelt, T., De ´chelotte, P., 2008. Autoantibodies
against appetite-regulating peptide hormones and neuropep-
tides: putative modulation by gut microflora. Nutrition 24,
Fetissov, S.O., Harro, J., Jaanisk, M., Ja ¨rv, A., Podar, I., Allik, J.,
Nilsson, I., Sakthivel, P., Lefvert, A.K., Ho ¨kfelt, T., 2005. Auto-
antibodies against neuropeptides are associated with psycholo-
gical traits in eating disorders. Proc. Natl. Acad. Sci. U. S. A. 102,
Guijarro, A., Laviano, A., Meguid, M.M., 2006. Hypothalamic inte-
gration of immune function and metabolism. Prog. Brain Res.
Hahn, T.M., Breininger, J.F., Baskin, D.G., Schwartz, M.W., 1998.
Coexpression of Agrp and NPY in fasting-activated hypothalamic
neurons. Nat. Neurosci. 1, 271—272.
Harris, J.I., Lerner, A.B., 1957. Amino-acid sequence of the alpha-
melanocyte-stimulating hormone. Nature 179, 1346—1347.
Heisler, L.K., Cowley, M.A., Tecott, L.H., Fan, W., Low, M.J., Smart,
J.L., Rubinstein, M., Tatro, J.B., Marcus, J.N., Holstege, H., Lee,
C.E., Cone, R.D., Elmquist, J.K., 2002. Activation of central
melanocortin pathways by fenfluramine. Science 297, 609—611.
148M.H. Sinno et al.
Kas, M.J.H., Bruijnzeel, A.W., Haanstra, J.R., Wiegant, V.M., Adan,
R.A.H., 2005. Differential regulation of agouti-related protein
and neuropeptide Yin hypothalamic neurons following a stressful
event. J. Mol. Endocrinol. 35, 159—164.
Kishi, T., Elmquist, J.K., 2005. Body weight is regulated by the
brain: a link between feeding and emotion. Mol. Psychiatry 10,
Kuo, L.E., Kitlinska, J.B., Tilan, J.U., Li, L., Baker, S.B., Johnson,
M.D., Lee, E.W., Burnett, M.S., Fricke, S.T., Kvetnansky, R.,
Herzog, H., Zukowska, Z., 2007. Neuropeptide Y acts directly
in the peripheryon fat tissue and mediates stress-induced obesity
and metabolic syndrome. Nat. Med. 13, 803—811.
Lang, B., Dale, R.C., Vincent, A., 2003. New autoantibody mediated
disorders of the central nervous system. Curr. Opin. Neurol. 16,
Liu, J., Garza, J.C., Truong, H.V., Henschel, J., Zhang, W., Lu, X.Y.,
2007. The melanocortinergic pathway is rapidly recruited by
emotional stress and contributes to stress-induced anorexia
and anxiety-like behavior. Endocrinology 148, 5531—5540.
Marsh, D.J., Hollopeter, G., Huszar, D., Laufer, R., Yagaloff, K.A.,
Fisher, S.L., Burn, P., Palmiter, R.D., 1999. Response of melano-
cortin-4 receptor-deficient mice to anorectic and orexigenic
peptides. Nat. Genet. 21, 119—122.
Maslova, L.N., Bulygina, V.V., Markel, A.L., 2002. Chronic stress
during prepubertal development: immediate and long-lasting
effects on arterial blood pressure and anxiety-related behavior.
Psychoneuroendocrinology 27, 549—561.
Meerlo, P., Horvath, K.M., Nagy, G.M., Bohus, B., Koolhaas, J.M.,
1999. The influence of postnatal handling on adult neuroendo-
crine and behavioural stress reactivity. J. Neuroendocrinol. 11,
a family of genes that encode the melanocortin receptors.
Science 257, 1248—1251.
S.N., Numa, S., 1979. Nucleotide sequence of cloned cDNA for
bovine corticotropin-beta-lipotropin precursor. Nature 278,
Pellow, S., Chopin, P., File, S.E., Briley, M., 1985. Validation of
open:closed arm entries in an elevated plus-maze as a measure
of anxiety in the rat. J. Neurosci. Methods 14, 149—167.
Peter, J.-C., Nicholson, J.R., Heydet, D., Lecourt, A.-C., Hoebeke,
J., Hofbauer, K.G., 2007. Antibodies against the melanocortin-4
receptor act as inverse agonists in vitro and in vivo. Am. J.
Physiol. Regul. Integr. Comp. Physiol. 292, R2151—R2158.
Schio ¨th, H.B., 2001. The physiological role of melanocortin recep-
tors. Vitam. Horm. 63, 195—232.
Sergeyev, V., Fetissov, S., Mathe, A.A., Jimenez, P.A., Bartfai, T.,
Mortas, P., Gaudet, L., Moreau, J.L., Ho ¨kfelt, T., 2005. Neuro-
peptide expression in rats exposed to chronic mild stresses.
Psychopharmacology (Berl) 178, 115—124.
Steinman, L., 2004. Elaborate interactions between the immune and
nervous systems. Nat. Immunol. 5, 575—581.
Sun, H.D., Malabunga, M., Tonra, J.R., DiRenzo, R., Carrick, F.E.,
Leibel, R.L., Kussie, P., 2007. Monoclonal antibody antagonists of
hypothalamic FGFR1 cause potent but reversible hypophagia and
weight loss in rodents and monkeys. Am. J. Physiol. Endocrinol.
Metab. 292, E964—E976.
Swanson, L.W., 1998. Brain Maps: Structure of the Rat Brain, second
ed. Elsevier Science B.V., Amsterdam.
the site of anorectic action of peripherally administered synthetic
melanocortin peptide MT-II in rats. Brain Res. 977, 221—230.
Wilson, J.F., Morgan, M.A., 1980. Plasma concentrations of alpha-
melanotropin in the rat during the acquisition and extinction of
conditioned avoidance behaviour and during the acquisition of
maze learning behaviour. Psychopharmacology (Berl) 68, 67—72.
Zelena, D., Haller, J., Halasz, J., Makara, G.B., 1999. Social stress of
variable intensity: physiological and behavioral consequences.
Brain Res. Bull. 48, 297—302.
Zorrilla, E.P., Iwasaki, S., Moss, J.A., Chang, J., Otsuji, J., Inoue, K.,
Meijler, M.M., Janda, K.D., 2006. From the cover: vaccination
against weight gain. Proc. Natl. Acad. Sci. U. S. A. 103, 13226—
MSH autoantibodies in feeding and anxiety149