The project was not funded commercially, although the authors received a gift of the antagonist compound JMV2959 from AeternaZentaris GMBH, Frankfurt, Germany. This does not alter the authors' adherence to all the PLOS ONE policies on sharing data and materials.
Conceived and designed the experiments: KPS EE SLD ZL IF MAC EH. Performed the experiments: MAC EH CSM IF. Analyzed the data: MAC EH IF. Wrote the paper: MAC SLD.
Here, we sought to demonstrate that the orexigenic circulating hormone, ghrelin, is able to exert neurobiological effects (including those linked to feeding control) at the level of the amygdala, involving neuroanatomical, electrophysiological and behavioural studies. We found that ghrelin receptors (GHS-R) are densely expressed in several subnuclei of the amygdala, notably in ventrolateral (LaVL) and ventromedial (LaVM) parts of the lateral amygdaloid nucleus. Using whole-cell patch clamp electrophysiology to record from cells in the lateral amygdaloid nucleus, we found that ghrelin reduced the frequency of mEPSCs recorded from large pyramidal-like neurons, an effect that could be blocked by co-application of a ghrelin receptor antagonist. In
The stomach-derived hormone, ghrelin
Another key brain area linking feeding control with emotional reactivity/mood is the amygdala. The amygdala is involved in memory and emotion processing but it also plays an important role in the modulation of reward, learning, and attention
Studies to date linking ghrelin with emotional reactivity/mood (especially anxiety-like behavior) are not altogether in agreement. For example, acute ghrelin injection has been shown to both increase
In the present study, we explored the amygdala as a target for ghrelin's behavioral effects linked to appetite control and emotional reactivity/mood. Given the uncharted nature of the amygdala GHS-R distribution, we first sought to describe the presence and neuroanatomical distribution of this receptor in the amygdala of the rat. Next, we explored cellular electrophysiological responses of amygdala cells to ghrelin and ghrelin antagonists. Finally, we determined the effects of intra-amygdala ghrelin injection on food intake and also on anxiety-like behavior in both the EPM and open field tests, that takes into account food availability immediately prior to the tests.
All these studies were carried out with ethical permissions from the Animal Welfare Committee (IEM, Budapest) and from the Animal Ethics committee at the University of Gothenburg and in accordance with legal requirements of the European Community.
For electrophysiology and
For studies investigating food intake and anxiety-like behaviour, adult male Sprague-Dawley rats (200–250 g, Charles River, Sulzfeld, Germany) were used. All animals were housed in a 12 hr light/dark cycle and maintained under controlled temperature (20–22°C) with free access to regular chow and water, except where indicated otherwise.
Ghrelin (Tocris, Bristol, UK) was dissolved in physiological saline. For food intake and anxiety studies, the dose of ghrelin selected for intra-amygdala injection was 0.3 and 1 µg/rat, microinjected in a volume of 0.2 µl saline. The 1 µg dose was used previously to drive food intake when administered to the ventral tegmental area or nucleus accumbens, while the lower dose was sub-threshold
Here we sought to provide a clear account of the distribution of GHS-R in the amygdala of the rat. Pre-hybridization, hybridization and post-hybridization procedures were performed on slide-mounted 12 µm frozen sections
Rats were anesthetized using isoflurane inhalation. The brain was removed rapidly and immersed in ice-cold cutting solution (in mM: sacharose 252.0, KCl 2.5, NaHCO3 26.0, CaCl2 2.0, MgCl2 2.0, NaH2PO4 1.25, and glucose 10) bubbled with a mixture of 95% O2 and 5% CO2. 400-µm-thick coronal slices were then prepared containing the lateral amygdala with a VT-1000S vibratome (Leica GmBH, Wetzlar, Germany) in ice-cold oxygenated cutting solution. The slices were bisected along the midline and transferred into artificial cerebrospinal fluid (aCSF, in mM: NaCl 126.0, KCl 2.5, NaHCO3 26.0, CaCl2 2.0, MgCl2 2.0, NaH2PO4 1.25, and glucose 10) saturated with O2/CO2, and kept in it for 1 hr to equilibrate. Equilibration started at 33°C and was allowed to cool to room temperature. Electrophysiological recordings were carried out at 33°C, during which the brain slices were oxygenated by bubbling the aCSF with O2/CO2 gas. Axopatch 200B patch clamp amplifier, Digidata-1322A data acquisition system, and pCLAMP 9.2 software (Molecular Devices Co., Sunnyvale, CA) were used for recording. Cells were visualized with a BX51WI IR-DIC microscope (Olympus Co., Tokyo, Japan). The patch electrodes (OD = 1.5 mm, thin wall; Garner Co., Claremont, CA) were pulled with a Flaming-Brown P-97 puller (Sutter Instrument Co., Novato, CA) and polished with an MF-830 microforge (Narishige, Tokyo, Japan). After control recording, the slices were treated with various drugs for 10 min and the recording repeated for 250 sec. Each neuron served as its own control when effects of the drugs were evaluated.
The cells were voltage clamped at a −70 mV holding potential. The pipette solution contained (in mM): HEPES 10.0, K-gluconate 136.0, KCl 4, EGTA 5.0, CaCl2 0.1, Mg-ATP 4.0, and Na-GTP 0.4 (pH 7.3 with NaOH). The resistance of the patch electrodes was 3–5 MΩ. Spike-mediated transmitter release was blocked in all experiments by adding the voltage-sensitive Na-channel inhibitor tetrodotoxin (TTX, 750 nM; Tocris) to the aCSF 10 min before control miniature excitatory postsynaptic currents (mEPSCs) were recorded. Picrotoxin (100 µM; Sigma, St. Louis, MO) was also used in all experiments in the aCSF to block GABAA-receptor mediated inhibitory postsynaptic currents. After establishing whole-cell clamp configuration the mEPSCs were recorded. After this, rat ghrelin (4 µM) or the ghrelin receptor antagonist JMV2959 (10 µM) were applied for 10 min in the aCSF and the mEPSCs were recorded.
The rats were acclimatized to the animal facility for at least 7 days before commencing surgery. They were anesthetized with isoflurane and placed in a stereotaxic apparatus for catheter placement. A stainless steel guide cannula (26 gauge; Plastics One, Roanoke, VA, USA) was positioned in the amygdala using the following stereotaxic coordinates from bregma: anteroposterior – 2.8 mm, lateral 4.8 mm, dorsoventral – 6.6 mm), according to a rat brain atlas
A: Photomicrograph of a 40 µm counterstained coronal section of rat brain at level Bregma −3.3, illustrating the injection site. B: Schematic representation of the amygdala according to the rat brain atlas
These experiments were designed to determine whether ghrelin action at the level of the amygdala is important for food intake in rats. Ghrelin (0.3 or 1 µg/rat) or vehicle were injected unilaterally into the amygdala of satiated rats and food intake measured at 1, 2, 4 and 24 hr after injection. A counterbalanced design was used where each rat received all the possible treatments, saline, ghrelin 0.3 µg and ghrelin 1 µg, each separated by a minimum of 48 hr. Using an identical experimental design, food intake was also measured after intra-amygdala injection of JMV2959 (2 or 10 µg/rat) or vehicle to rats that had been fasted overnight.
The experimental protocol is illustrated schematically in
Fed rats were injected with saline or ghrelin directly into the amygdala at time zero. In the FOOD ACCESS paradigm, rats were allowed access to food during the first hour after injection whereas food access was denied in the FOOD WITHHELD paradigm. After this, all rats underwent tests exploring anxiety-like behaviour, first in the EPM test (5 min) and then in the open field test (40 min). Afterwards all the rats were returned to their home cages and post-test food intake measured for 1 hr (corresponding to time 2–3 hr after injection).
The EPM apparatus consisted of two open arms (50×10 cm2) made of black PVC (polyvinyl chloride), crossed by two closed arms of the same size but with walls of 40 cm high. All the arms emerged from a central platform placed 50 cm above the floor. Under dim light (around 100 lux over the open arms and 60 lux over the closed arms) every rat was placed in the central platform and they were allowed to freely move in the whole apparatus during 5 min. The EPM apparatus was cleaned between each trial. The rat behavior was recorded by a video camera during this time for subsequent determination of the following parameters: the number of entries in the open arms, the number of entries into the closed arms (an entry was counted when the four paws were placed on the respective arm) and the time spent in the open arms and in the closed arms.
At the end of EPM testing, the rats were undisturbed for 10 min before being placed in sound attenuated locomotor activity boxes (40 min) for the open field test. These boxes, measuring 1×1×0.5 m contained an inner Plexiglas chamber (0.7×0.7×0.35 m) and they were equipped with an automatic system integrated by two rows of eight photocells on each side, that allow to detect movements (Kungsbacka Mät- och reglerteknik, Sweden). Under dim light (40 lux) the locomotor activity is assessed by registration of the breaking of a sequence of beams. The peripheral activity is measured by the photocells closed to the corners while the central activity is registered by the central 6 photocells on each side. The photocells in the upper row measure rearing. Finally they were returned to their home cage for further food intake measurement (taken during the 2–3rd hr after ghrelin/saline injection).
In the electrophysiology studies each experimental group contained 8–10 recorded cells from six to seven animals. Recordings (250 sec) were stored and analysed off-line. Event detection was performed using the Clampfit module of the PClamp 9.2 software (Molecular Devices Co.). Group data were expressed as mean ± SEM and percentage change in the frequency of the mEPSCs due to the application of the ghrelin or the JMV2959 was calculated. Statistical significance was analysed using Student's
For the behavioral experiments the data were analyzed in IBM SPSS Statistics 9. The “P" values <0.05 were considered statistically significant. All the data are presented as mean ± Standard Error of the Mean (SEM).
The autoradiographic detection of
The autoradiographic detection of the isotopic
Neurons of the lateral amygdala could be categorized into two basic cell types: large, pyramidal-like neurons and smaller, non-pyramidal-like neurons
A) Application of ghrelin (4 µM) in the extracellular solution decreased the frequency of the mEPSCs. B) Extracellular administration of the ghrelin receptor antagonist JMV2959 (10 µM), blocked this effect of ghrelin. C) Histogram shows the relative percentages of mEPSC frequencies. *P<0.05.
In satiated rats, unilateral administration of the higher ghrelin dose (1 µg) to the amygdala induced a >2.5 fold increase in food intake relative to saline-treated controls, measured at 1, 2, and 4 hr after the microinjection. The lower dose (0.3 µg) was without effect (
Normal chow intake was measured at 1, 2, 4, and 24 hr following unilateral intra-amygdala injection of either (A) ghrelin (0.3 and 1 µg) to freely-fed rats or (B) JMV2959 (2 or 10 µg) to overnight fasted rats. Data are expressed as mean ± S.E.M. *P<0.05 **P<0.01 ***P<0.001, vs. saline. Paired sample t-test, SPSS.
First, we used the EPM test to explore the effects of intra-amygdala injection of ghrelin on emotional reactivity (anxiety-like behavior) in rats allowed access to food during the first hour following injection (FOOD ACCESS paradigm;
In rats given access to food during the first hour after intra-amygdala injection (FOOD ACCESS), ghrelin increased food intake relative to saline controls (g of chow), both during this hour and during the 1 hr measurement taken after the anxiety tests (A). In this paradigm there was no effect of ghrelin (relative to saline controls) on anxiety-like behavior in either the EPM test (time spent in the open arm; B) or the open field test (central activity or central rearing; C, D respectively). *P<0.05 **P<0.01, vs. saline. Independent samples t-test, SPSS.
In this “FOOD WITHHELD" paradigm rats were denied access to food during the first hour after intra-amygdala injection. (A) An orexigenic response to intra-amygdala ghrelin injection was detected when animals were returned to their home cages after the anxiety testing. Intra-amygdala ghrelin injection decreased anxiety-like behavior relative to saline controls, reflected by an increase in the amount of time spent in the open arms in the EPM test (B) and by the increase in central activity (C) and central rearings (D) in the open field test. *P<0.05 **P<0.01, ***P<0.001, vs. saline. Independent samples t-test, SPSS.
The effects of intra-amygdala ghrelin injection on anxiety-like behavior were further explored/validated in the open field test: ghrelin increased both central activity and central rearing in the FOOD WITHHELD paradigm (
In the present study we explored the effects of ghrelin on the electrophysiological activity of amygdaloid pyramidal neurons and extended these findings to behavioral outputs of amygdala: food intake and anxiety-like behavior. First we validated the amygdala as a target for ghrelin by demonstrating the presence of ghrelin receptor (GHS-R) mRNA selectively in the lateral and medial amygdaloid nuclei of the rat. Using this information we performed whole cell patch clamp studies from amygdala cells located in the lateral amygdala, where GHS-R is most abundant. We found that ghrelin suppressed the electrical activity (i.e. a decreased number of mEPSCs) of pyramidal-like neurons in this region. Again, guided by localization of GHS-R, we demonstrated that intra-amygdala ghrelin injection (via catheters directed towards areas where GHS-R is most abundant) induced a robust feeding response. These data suggest that ghrelin's neurobiological effects in this brain area are linked to food intake, as shown previously for almost all other GHS-R-expressing brain areas studied, as reviewed previously
To our knowledge, this is the first study to describe the presence of GHS-R mRNA in the amygdala of the rat, by
Previous studies have shown that hyper-excitability of pyramidal-like neurons in the lateral amygdaloid nucleus is linked to anxiety-like behavior
Previous studies determining the effects of intra-amygdala injection of ghrelin on feeding behavior are not altogether in agreement; whereas one study reported no effect on regular chow intake
Previously we demonstrated that central administration of a ghrelin receptor antagonist suppresses the hyperphagia observed after an overnight fast
The source of endogenous ghrelin in the amygdala remains an open question. The afferent ghrelin signal to the amygdala could be blood-borne
For survival, it seems rather logical that appetite-regulating peptides would also regulate neurobiological pathways involved in emotional reactivity/mood. Consistent with this, anorexigenic peptides such as leptin
In the present study, we sought to clarify the neurobiological effects of ghrelin on anxiety-like behavior focusing only on those exerted at the level of the amygdala, that we now know are relevant for feeding control and likely linked to emotional reactivity. We reasoned that the outcome of the anxiety testing could be dependent upon food availability when the ghrelin is administered (i.e. whether animals are able to eat or not after receiving the ghrelin injection). In other words, the afferent ghrelin hunger signal is expected to coordinate a feeding response but if no food is available, it could be advantageous for survival if emotional (anxiety-like) behaviors that would otherwise limit the animal from finding food are suppressed. To explore these emotional behaviors, we focused especially on anxiety-like behavior (in the EPM test and the open field) for which an advantageous decrease in anxiety-like behavior is signified by an increased time spent in the open arm or the centre field respectively. Thus, we performed the anxiety tests 1 hr after intra-amygdala saline/ghrelin injection in two different experimental paradigms, one in which the animals were allowed to feed between the injection and the anxiety test and another in which food access was denied. In the EPM test, ghrelin increased both the number of entries into the open arm and the amount of time spent there (relative to saline-injected controls), in rats that were not given access to food during the first hour after the injection. Conversely, in rats allowed to feed during the first hour after intra-amygdala injection, ghrelin had no effect on any parameter measured in the EPM test. These effects of intra-amygdala ghrelin on anxiety-like behavior were further validated in the open field test. Additionally, we were able to show that overall locomotor activity was similar for saline-injected and ghrelin-injected rats, both those with access to food and in those denied access to food during the hour after injection; thus any differences observed in these tests are not secondary to a locomotor effect of ghrelin in the amygdala. Thus, ghrelin's role at the level of the amygdala appears to include the suppression of anxiety-like behavior (including those that limit the rat from finding food) when food is not available. In the present study we did not determine whether food availability per se alters emotional (anxiety-like) behavior in the anxiety tests. Thus, in situations when food is not available, further studies would be required to determine whether ghrelin reduces anxiety-like behavior per se or whether it prevents an anxiogenic effect of withholding food.
Given the electrophysiological and neuroanatomical data identifying the amygdala as a target for ghrelin, it seems rather likely that the behavioral effects observed after intra-amygdala ghrelin administration involve a direct action at the level of the amygdala. We know very little about local concentrations of ghrelin after parenchymal injection to the amygdala, or about its half-life when delivered via this route. It is theoretically possible that ghrelin could even diffuse to other brain areas that could be important for the observed behavioral effects, such as hypothalamus or closer structures such as the hippocampus. In previous studies, however, involving parenchymal injection of ghrelin to a different brain area, the ventral tegmental area, no effects were observed when the peptide was delivered “off-target" to neighboring structures using doses similar to those in the present study
In summary, our results provide neuroanatomical, electrophysiological and behavioral data indicating that the amygdala, especially the lateral nucleus (shown here to be the region with highest expression of the ghrelin receptor, GHS-R) is a key brain target for ghrelin, integrating effects of food intake and emotional reactivity. Our observation that ghrelin's acute effects to suppress anxiety-like behavior in the EPM test, exerted at the level of the amygdala, are only observed in rats that are prohibited from eating during the first hour after ghrelin administration, support the idea that ghrelin's effects on food intake and emotional reactivity are linked and likely facilitate exploration, foraging and other food-linked behaviors that are critical for survival when food is scarce.
We thank Jakob Näslund and Dr. Staffan Nilsson for their kind help with the anxiety tests and the statistics, respectively.