inhibitory circuits interacting with orexinergic neurons influence differentially feeding behaviors in hamsters

Behavioural Brain Research 234 (2012) 91–99

Contents lists available at SciVerse ScienceDirect

Behavioural Brain Research journal homepage: www.elsevier.com/locate/bbr

Research report

Amygdalar excitatory/inhibitory circuits interacting with orexinergic neurons influence differentially feeding behaviors in hamsters E. Avolio a,b , R. Alò a , M. Mele a , A. Carelli a , A. Canonaco b , L. Bucarelli a , M. Canonaco a,∗ a b

Comparative Neuroanatomy Laboratory of Ecology Department, University of Calabria, Ponte Pietro Bucci 4b, 87030 Arcavacata di Rende, Cosenza, Italy Health Center srl, Biomedical and Nutritional Center, via Sabotino 66, 87100 Cosenza, Italy

h i g h l i g h t s     

Increased food consumption following infusion of hamster BlA with ORX-A. Treatment of CeA with ORX-B prevailed on water consumption. ORX-A/-B + NMDA/zolpidem increased or reduced eating/drinking behaviors, respectively. The same treatments accounted for up and down ORX-2R expression levels. ORX-A/B + combined treatment support new therapeutic bearings on psychiatric disorders.

a r t i c l e

i n f o

Article history: Received 7 June 2012 Accepted 13 June 2012 Available online 20 June 2012 Keywords: Amygdala Feeding Hamster Orexin Glutamate GABA

a b s t r a c t Recently, environmental stimuli on different neurobiological events, via participation of distinct amygdalar (AMY) ORXergic fibers have aroused wide interests in view of their ability to modify neuronal linked stressful and physiological homeostatic conditions. Results of the present study indicate that ORXergic (ORX-A/B) circuits of the facultative hibernating golden hamster (Mesocricetus auratus) central AMY (CeA) and basolateral AMY (BlA) nuclei constitute major sites of feeding behaviors. Indeed, hamsters after treatment of BlA with ORX-A frequently ingested greater quantities of food as compared to controls, while ORX-B in CeA induced a very (p < 0.001) great consumption of water. The same nuclei treated separately with either ORX-A or ORX-B ± the selective ␣1 GABAA benzodiazepine receptor agonist (zolpidem) dedicated less time to eating and drinking sessions. Conversely, hamsters that received the same neuropeptides but this time with the glutamatergic agonist NMDA displayed greater hyperphagic effects above all for ORX-A. When behavioral changes were compared to the expression of the specific ORXergic receptor (ORX-2R), an up-/down-regulating pattern was detected in some limbic areas (AMY, hippocampus and hypothalamus) following treatment with ORX-A or ORX-B plus NMDA. Overall, indications deriving from this study strongly point to hamster BlA-enriched ORX-A fibers in combination with either inhibitory or excitatory signals as main targets of hyperphagic responses while CeA ORX-B activities in presence of these same neuronal signals predominantly induced drinking motivational behaviors. The distinct behavioral activities of these two neuropeptides may have useful clinical bearings toward psychiatric and sleeping disorders such as bulimia and narcolepsy. © 2012 Elsevier B.V. All rights reserved.

1. Introduction The two excitatory neuropeptides hypocretins or orexins (ORXA, B) that are produced from a single mRNA transcript plus pre-pro-hormone of the lateral hypothalamic area (LHA) seem to be capable of interacting with two receptor subtypes, i.e. ORX1R and ORX-2R. In particular ORX-1R binds more selectively with ORX-A while ORX-2R has an equal affinity for both ORX-A and

∗ Corresponding author. Tel.: +39 984 492974; fax: +39 984 492986. E-mail address: [email protected] (M. Canonaco). 0166-4328/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbr.2012.06.013

ORX-B [43]. Hypothalamic ORXergic circuits have demonstrated to be actively involved with feeding habits plus altering anxiety states, motor behaviors and sleep-wakening states [5]. Deletion of the pre-pro-orexin gene produced a narcoleptic-like phenotype in mice, which is similar to that of humans [10]. At the same time, intracerebroventricular (i.c.v.) administration of ORX-A accounted for a large number of LHA ORXergic neuronal fields promoting an appetite-stimulating type of activity in non-fasted rats [43]. In addition, ORXergic fibers of this hypothalamic area by cross-talking with the cerebral cortex (COR), hippocampus (HIP), septum, and amygdala (AMY) appear to control the different motivational and motor performances [28].

92

E. Avolio et al. / Behavioural Brain Research 234 (2012) 91–99

At the brain level, it has already been shown that some AMY nuclei are not only actively involved with the modulation of negative affective states such as fear [25] but above all with the promotion of appetitive processes, including feeding behaviors linked to reward tasks [21]. From a morpho-functional point of view, AMY exhibits distinct and segregated functional domains [44] that include a posterodorsal “cortical-like” basolateral amygdala nucleus (BlA) and a central amygdala nucleus (CeA), both of which are connected to hypothalamic and brainstem autonomic systems [49]. Recent works have reported that both AMY sites serve as major feeding input regions receiving and sending neuronal signals to LHA and nucleus accumbens shell (nAc) [39]. In the case of CeA, it serves as an output relay for intra-amygdaloid connections originating from BlA and so tends to emphasize the participation of these afferent fibers, which are innervated by gustatory signals [32] on feeding responses [55]. It is important to note that feeding behaviors depend, aside ORXergic signals, on the interaction of other neuroreceptor systems such as dopamine, opioids, glutamate and ␥-amino-butyric acid (GABA) as shown by their role on food seeking and consumption plus feeding reward processes [22]. Works have shown that certain subpopulations of GABAA receptors (GABAA Rs), characterized by specific subunit compositions exert critical effects on feeding and motor activities [6]. This supramolecular pentameric structure, composed of at least 20 different classes of subunits: ␣ (1–6), ␤(1–4), ␥ (1–3), ␦, ␧, ␪, ␲, ␳ (1–3), forms a complex GABAA R ionophore molecule [36]. Due to the involvement of the ␣ subunit with the structural assembly of this neuroreceptor system plus the expression of pharmacological functions as shown by ␣1,2,3,5 exhibiting varying degrees of sensitivity to benzodiazepines (BZD) underlies the importance of such a subunit on the execution of socio-sexual and motor behaviors [7]. Working along these lines of interest, the differing ␣-containing GABAA R neuronal circuits represent a major facet of appetitive processes, including food reward behaviors [22]. In particular telencephalic GABAergic activities deriving from the nAc shell have been shown to influence hyperphagic effects [41]. On the basis of the above features, it was the intention of the present study to establish the type of feeding responses induced by separate i.c.v. infusions of BlA and CeA with the two ORX neuropeptides (ORX-A; ORX-B) as well as together with either the BZD agonist zolpidem, specific for ␣1 GABAA R subunit, or the glutamate agonist N-methyl-D-aspartate (NMDA) in the facultative Syrian hibernating hamster (Mesocricetus auratus). The selection of this rodent was largely based on its permissive facultative hibernating features, which allow us to correlate motor performances to the various feeding requirements during certain stages of this physiological state. Hibernation is a unique physiological state that permits animals to survive under extraordinary climatic and stressful conditions. This condition has been largely studied on M. auratus that displays profound decreases in oxidative metabolism and body temperature during bouts of prolonged torpor interrupted every 5–14 days by brief periodic arousals. The different physiological states feature a decrease in body temperature, with short periods of inter-bout euthermy when body temperature rises to ∼37 ◦ C and is maintained for 12–48 h before reentry into torpor [12]. Indeed, the decrease in body temperature to values <5 ◦ C contributes to the reduction of metabolic and enzymatic activities reaching 2–4% of normal rates, which are restored rapidly to near-normal levels so to avoid complications prior to arousal state [47,48]. In the case of AMY sites, BlA and CeA have shown to be tightly linked with the onset of explorative and feeding type of behaviors [1]. For this reason the behavioral effects were compared to the expression differences of the main ORX receptor subtype (ORX-2R) in some limbic areas and precisely BlA, CeA, hippocampal dentate gyrus (DG), lateral amygdalar nucleus (Lat), LHA the periventricular

hypothalamic nucleus (Pe), the supraoptic nucleus (SON) that have been recognized as key ORXergic sites controlling different motor performances [33]. In addition, ORX-2R was preferred for this study due to its comparable affinity for both neuropeptides with respect to the greater affinity of ORX-1R for mostly ORX-A [42] plus to its importance on behavioral plasticity events [52]. The prevailing and distinct effects exerted by these neuropeptides may supply further insights regarding AMY-related feeding activities especially during some psychiatric disorders such as anorexia and bulimia, which are characterized by a disrupted motivational state [38]. 2. Materials and methods 2.1. Animals and surgery All experimental procedures described below were approved by the local Committee for Ethics in Animal Research (CEUA-UFSC, protocols #PP00091/CEUA and 23080.010535/2007-26) plus in accordance with suggestions and indications provided by the “Principles of animal care” (NIH, 1985). For the present study adult Syrian facultative hibernating hamsters (150–180 g body weight; Charles River, Como Italy) that were initially housed not more than 3 per cage were maintained at 22–24 ◦ C, under a 14:10 light–dark cycle (lights on 6 a.m.) with free access to food and water. After 1 week of habituation period, hamsters were anesthetized with ketamine hydrochloride (100 mg/kg, i.p.) and xylazine (20 mg/kg, i.p.) and some animals were stereotaxically implanted with a unilateral stainless steel guide cannula (30 G) aimed 2 mm dorsal to the CeA (n = 47) while other animals were implanted for treatment of BlA (n = 47), according to previously described studies [5]. The cannula was positioned onto the skull with jeweler screws, fixed with dental cement and assured during the entire experiment by an inner removable stylet (Fig. 1).

2.2. Drugs and injection ORX-A (20 nM), ORX-B (60 nM), zolpidem (100 nM) and NMDA (0.1 mM) all purchased from Sigma Chemical Co. (St. Louis, MO, USA) were freshly dissolved in saline 0.9%. Injections were made through an inner cannula (33 G) that extended 2 mm beyond the tip of the guide cannula, which was connected to a Hamilton microsyringe (1 ␮l) by polyethylene tubing. Injection volumes (1 ␮l) were given over a period of 60 s plus a further 60 s time-interval was used to consent the solution to diffuse from the cannula.

2.3. Experimental procedures Seven days after surgery, some free-feeding hamsters bearing CeA guide cannula received daily either doses of ORX-A (n = 5), ORX-B (n = 5), zolpidem (n = 5) and NMDA (n = 5) alone or in combination (ORX-A/B ± zolpidem, n = 10; ORXA/B ± NMDA, n = 10) for one week with respect to controls (C, n = 5) that received injections of 1 ␮l vehicle (0.9% saline). Other hamsters, this time bearing BlA guide cannula received the same above treatments. Immediately after injections, hamsters were initially placed for 30 min in the recording chamber to familiarize with its new environment. Subsequently, 30 min after the habituation period, three daily behavioral observations (10 a.m. to 6 p.m.) were conducted for the above treated hamsters during the entire observational session. Each day latency, duration and frequency of six previously defined behavioral categories and namely eating, drinking, grooming plus some typical motor performances such as moving about in the cage, following other hamsters, darting, upright position and resting [16], were continuously recorded by a webcam positioned perpendicularly at 60 cm above the cage floor, after which detailed behavioral analyses were conducted by using a Etholog 2.2 program [37]. At the end of the recording period, food that occasionally spilled onto the floor was recovered and weighed with food that remained on the feeder. The differences between food and water at beginning and end of recording period were calculated as total amount of food or water consumed. Also body weight differences of hamsters belonging to same treatment groups were compared to those of C at the end of the observation interval. In order to avoid interferences during behavioral recordings, back and lateral walls along with the floor cage were coated with a black adhesive plastic paper, while the front wall of the testing cage had a mirror of equal dimensions held at a 45◦ angle to the vertical plane so that it prevented animals from seeing its reflection.

2.4. Histological analysis At the end of behavioral procedures, hamsters were anesthetized and 1 ␮l Evans blue dye (1%) was injected to verify the correct sites of the cannula within either CeA (n = 2) or BlA (n = 2; Fig. 1). Brain sections (30 ␮m) obtained at a vibratome were stained with cresyl violet and cannula loci were established by using a camera lucida attached to a light microscope and then mapped onto corresponding schemes of the hamster atlas [30].

E. Avolio et al. / Behavioural Brain Research 234 (2012) 91–99

93

Fig. 1. Coronal brain schemes illustrating distinct CeA and BlA injections with ORX-A/-B alone or in combination with zolpidem/NMDA. In the schema, the cannula injection sites of the two AMY nuclei are indicated in cresyl violet stained histological sections showing the positioning of the cannula after having injected Evans blue dye. Stereotaxic coordinates were obtained from the hamster brain atlas. Abbreviations: HIP, hippocampus; AC, anterior commissure; BlA, basolateral amygdala nucleus; CeA, central amygdala nucleus; DG, hippocampal dentate gyrus; Lat, lateral amygdalar nucleus; LHA, lateral hypothalamic area; SON, supraoptic nucleus; Pe, periventricular hypothalamic nucleus.

2.5. In situ hybridization assay The effects of the same above combined treatments were carried out on other hamsters in order to establish the type of ORX-2R expression pattern occurring in the various limbic neuronal fields during the different behaviors. The brain of all treatment groups, after their final drug injection, was removed and processed for in situ hybridization procedures. For this part, ORX-2R antisense and sense probes were designed on the partial sequences of our rodent model and labeled by 3 tailing using digoxigenin-11-dUTP (DIG) according to indications supplied by DIG oligonucleotide tailing kit (Roche, Italy). Probes were prepared via their incubation at 37 ◦ C for 30 min and then stopped with 0.2 M EDTA pH 8.0. The concentrations applied were determined by its quantification against standards on Hybond N + filters (Amersham, Italy). Afterwards brain sections (10 ␮m), which were previously mounted on polylysine coated slides (Carlo Erba, Italy) and stored at −40 ◦ C, were incubated with 100 ng of antisense probe in 100 ␮l of hybridization solution for overnight incubation at 50 ◦ C in a humidified chamber [23]. Nonspecific hybridization was obtained on slides incubated with the sense probe. For immunological detection, sections were cover slipped for 45 min with PBS buffer containing 2% normal sheep serum (Sigma, Italy) and 0.3% Triton X100 (Sigma, Italy). Then an anti-digoxigenin alkaline phosphatase antibody (Roche, Italy) 1:100 was added for 2 h at room temperature and the alkaline phosphatase color reaction buffer (NBT/BCIP) was added to sections and incubated for 72 h in a humidified dark chamber. Hybridization signals were observed at a bright-field Dialux EB 20 microscope (Leitz) under a phase contrast objective (×40) and transcription activity was evaluated with a Panasonic Telecamera (Canon Objective Lens FD 50 mm, 1:3.5) attached to a Macintosh computer-assisted analyzer system by running an Image software of National Institutes of Health (Scion-Image 2.0) plus an internal standard curve for

calibrating O.D. values. O.D. density values were calculated on cresyl violet stained limbic nuclei identified using the hamster atlas [30]. 2.6. Data analysis All behavioral performances considered in the present study and namely eating, drinking, plus grooming along with body weight differences were expressed as a percentage (% ± s.e.m.) with respect to C. These behavioral changes were reported as the effects of ORX-A, ORX-B, ␣1 GABAA R agonist, glutamatergic agonist given alone or in combination (ORX-A/-B + ␣1 GABAA R agonist/glutamatergic agonist) with respect to C in either the AMY CeA or BlA via the application of a two-way ANOVA followed by Newman-Keul’s multiple range post hoc analysis when p < 0.05. Even the transcriptional capacities of some limbic neuronal fields to express ORX-2Rs (mean ± s.e.m.) were analyzed for these same treatments by using ANOVA followed by a Newman-Keul’s multiple range post hoc analysis with respect to C when p < 0.05.

3. Results 3.1. Feeding behaviors regulated by ORXs/GABA amygdalar interaction Treatment of the two major AMY brain areas (BlA and CeA) with daily microinjections of ORX-A or ORX-B alone as well as in combination with the BZD agonist zolpidem prior to behavioral

94

E. Avolio et al. / Behavioural Brain Research 234 (2012) 91–99

Fig. 2. Effects of ORX-A/-B ± zolpidem i.c.v. injections into BlA of hamsters (n = 5). For this part hamsters received a single infusion of following compounds: 1 ␮l/hamster of 20 nM ORX-A; 1 ␮l/hamster of 60 nM ORX-B; 1 ␮l/hamster of 100 nM zolpidem; 1 ␮l/hamster of ORX-A + zolpidem; 1 ␮l/hamster of ORXB + zolpidem and compared to C (animals infused with only saline solution). The test was performed 30 min after i.c.v. injections. Each bar represents % mean ± s.e.m. of feeding behaviors and body weight variations with respect to C that received saline solution. The differences were evaluated by ANOVA plus a post hoc Newman–Keul’s test, *,a p < 0.05, **,b p < 0.01 and ***,c p < 0.001. ‘*’ refers to significant differences with respect to controls; a,b,c refer to significant variations of different drug treatments with respect to either ORX-A or ORX-B treated hamsters.

observations proved to efficaciously modify feeding behaviors of the facultative hibernating Syrian hamster. Of all the behaviors considered and namely eating, drinking, grooming, rearing plus resting (data not shown for 3 latter behaviors), only the first two displayed significant variations and so our attention was only directed to them throughout the entire study. Indeed, hamsters BlA infused with ORX-A [F(3,17) = 5.03; p < 0.01] consumed notably (p < 0.001) elevated quantities of food (220%) with respect to C (Fig. 2A). This activity was further strengthened by an evident 20% increase in body weight caused by the neuropeptide, which turned out to be of a notably elevated entity (180%) when compared to C. Food consumption following treatment of BlA with ORX-A + zolpidem seemed to be inverted, despite this BZD agonist alone accounted for a elevated reduction of food consumption (−120%) with respect to animals treated with only the neuropeptide (Fig. 2A). Even body weight variations reflected substantial feeding changes as displayed by an elevated decrease in body weight (−160%) despite only great (p < 0.01) weight reductions were detected in animals treated with zolpidem with respect to C (−75%; Fig. 2A). Conversely when BlA was infused with the other neuropeptide (ORX-B), consistent feeding variations [F(3,17) = 3.18; p < 0.05] were observed (Fig. 2B). In particular, hamsters consumed moderate (p < 0.05) quantities of food (42%) while it was not altered following the addition of zolpidem in spite of this GABAA R agonist alone only accounting for a moderate reduction of food consumption (−35%). Regarding

Fig. 3. Effects of ORX-A/-B ± zolpidem i.c.v. injections into CeA of hamsters (n = 5). Even for this part, hamsters received a single infusion of following compounds: 1 ␮l/hamster of 20 nM ORX-A; 1 ␮l/hamster of 60 nM ORX-B; 1 ␮l/hamster of 100 nM zolpidem; 1 ␮l/hamster of ORX-A + zolpidem; 1 ␮l/hamster of ORXB + zolpidem and compared to controls. Each bar represents % mean ± s.e.m. of feeding behaviors and body weight variations with respect to controls. Differences were evaluated by ANOVA plus a post hoc Newman–Keul’s test, *,a p < 0.05, **,b p < 0.01 and ***,c p < 0.001. For asterisk and letter indications check Fig. 2.

drinking bouts, ORX-B did not appear to greatly influence them as shown by a 57% increase of water consumption (Fig. 2B). Similarly, weight changes were not of a consistent entity as exhibited by a great increase of weight (65%) and this variation resulted to be greatly reduced in the presence of zolpidem when compared to only ORX-B treatment (−75%). On the other hand, infusion of CeA nucleus with ORX-A seemed to moderately increase, if any, total food consumption [F(3,17) = 3.22, p < 0.05] aside the notably elevated increment of food intake (175%) detected after treatment with the neuropeptide while only a moderate increase (55%) of body weight was reported when compared to C, respectively (Fig. 3A). Following the addition of ␣1 GABAA agonist hamsters ate less even though in a non-significant fashion with respect to C whereas it seemed to account for a notably reduced amount of food intake (−130%) with respect to only ORXA-treated hamsters. This trend goes in the same direction of body weight changes as revealed by neither zolpidem alone or combined with ORX-A being able to modify the great consumption of the neuropeptide nonetheless their body weight resulted to be moderately reduced (−71%) with respect to hamsters treated with only the neuropeptide (Fig. 3A). Surprisingly, infusion of this AMY nucleus with ORX-B [F(3,17) = 5.31; p < 0.01] caused hamsters to consume elevated quantities of water (290%) while it led animals to only moderately consume food (57%) with respect to C (Fig. 3B). The levels of water consumption by CeA was further enhanced when ORX-B was given contextually with the ␣1 GABAA agonist as displayed by greatly reduced drinking rhythms with respect to C and this diminishing trend turned out to be of a notably great nature (−160%) with respect to ORX-B treatments (Fig. 3B).

E. Avolio et al. / Behavioural Brain Research 234 (2012) 91–99

95

Fig. 4. Effects of ORX-A/-B ± NMDA i.c.v. injections into BlA of hamsters (n = 5). Even for this part, hamsters received a single infusion of following compounds: 1 ␮l/hamster of 20 nM ORX-A; 1 ␮l/hamster of 60 nM ORX-B; 1 ␮l/hamster of 0.1 mM NMDA; 1 ␮l/hamster of ORX-A + NMDA; 1 ␮l/hamster of ORX-B + NMDA and compared to controls. Each bar represents % mean ± s.e.m. of feeding behaviors and body weight changes with respect to controls. Differences were evaluated by ANOVA plus a post hoc Newman–Keul’s test, *,a p < 0.05, **,b p < 0.01 and ***,c p < 0.001. For asterisk and letter indications check Fig. 2.

Fig. 5. Effects of ORX-A/-B ± NMDA i.c.v. injections into CeA of hamsters (n = 5). Even for this part, hamsters received a single infusion of following compounds: 1 ␮l/hamster of 20 nM ORX-A; 1 ␮l/hamster of 60 nM ORX-B; 1 ␮l/hamster of 0.1 mM NMDA; 1 ␮l/hamster of ORX-A + NMDA; 1 ␮l/hamster of ORX-B + NMDA and compared to controls. Each bar represents % mean ± s.e.m. of feeding behaviors and body weight changes with respect to controls. Differences were evaluated by ANOVA plus a post hoc Newman–Keul’s test, *,a p < 0.05, **,b p < 0.01 and ***,c p < 0.001. For asterisk and letter indications check Fig. 2.

3.2. Effects of NMDA injection on ORXs in feeding behaviors

food consumed when the AMY nucleus received ORX-A (175%) with respect to C continued to be even greater in the presence of the glutamatergic agonist (320%), despite such an increase resulted to be of only a great entity (68%) when it was compared to the effects induced by only the neuropeptide. The efficacious feeding actions of NMDA on ORX-A were further supported by increased body weight although it only turned out to be of a great nature with respect to ORX-A-treated hamsters (Fig. 5B). Conversely, NMDA did not modify ORX-B-dependent eating and drinking effects [F(3,17) = 4.95, p < 0.01] as shown by comparable feeding and drinking events with respect to ORX-B alone (Fig. 5B). In addition, the great NMDAdependent effects were not able to modify ORX-B-related weight changes.

In the case of both neuropeptides being administered with NMDA, it was possible to observe a synergic effect of this excitatory ionotrophic factor [F(3,17) = 5.11, p < 0.01] above all when infused together with ORX-A into BlA. Indeed hamsters treated with ORXA consumed notably great amounts of food (220%) that still turned out to be of a notable great nature, despite it being somewhat numerically greater following the addition of NMDA (83%) with respect to ORX-A-treated animals (Fig. 4A). Even body weight displayed a similar pattern as demonstrated by a notably elevated increase in body weight (245%) against C, which instead turned out to be of a moderate type (48%) with respect to the notably great increases of animals treated with only ORX-A. On the other hand, BlA that received ORX-B did not exert the same behavioral performances as shown by only moderately differing feeding and drinking events [F(3,17) = 3.15; Fig. 4B]. In this case moderate food consumption evoked by the neuropeptide alone proved to be only greatly enhanced (72%) when combined with NMDA with respect to C, effect which then proved to be of a notably great entity when compared to ORX-B alone (115%). Conversely, the major effects of NMDA on body weight did not further modify those promoted in the presence of ORX-B with respect to the neuropeptide alone (Fig. 4B). For the treatments of CeA, it was still feeding behaviors that turned out to be greatly influenced [F(3,17) = 5.25; p < 0.01] by ORXA and NMDA (Fig. 5A). In this context, the evident quantities of

3.3. In situ hybridization In situ hybridization analysis displayed an evident heterogeneous distribution pattern of ORX-2R expressing neurons as indicated by the differing expression levels of this receptor in a representative photogram of posterior brain areas (Fig. 6a). The same above treatments appeared to evoke distinct ORX-2R expression patterns in some limbic areas as shown by dense ORX2R-containing neurons areas labeled with antisense probe (Fig. 6a and b) compared to light non-labeled levels in presence of the sense probe (Fig. 6a and b ). In a first case, the addition of ORX-A to BlA accounted for a great (65–72%) up-regulation of ORX-2R densities in the two

96

E. Avolio et al. / Behavioural Brain Research 234 (2012) 91–99

Fig. 6. The heterogeneous distribution pattern of ORX-2R expressing neurons in a representative hamster posterior forebrain sections (a, a ). Expression levels of ORX-2R appeared to be highly specific as shown by the very evident and heterogeneous labeling pattern of ORX-2R-expressing neurons with respect to the almost background levels of the sense probe (b, b ). For abbreviations check Fig. 1.

hypothalamic sites (LHA, Pe) involved with feeding behaviors, whereas only a moderate (40%) increase was detected for Lat (Fig. 7a). The up-regulating effects of ORX-A were still further increased when the neuropeptide was combined with NMDA as exhibited by the notably elevated and enhanced ORX-2R mRNA levels in LHA (92%) and in Pe (68%) plus Lat (62%), respectively, when compared to C, in spite of the lack of any further significant variations of the latter two hypothalamic nuclei when compared to ORX-A treatment alone. In addition NMDA effects, instead, turned out to be of an inhibitory type in CeA as demonstrated by the moderate reduction (−37%) of ORX-2R levels with respect to C; a trend which appeared to be very greatly reduced (−147%) when compared to ORX-A-treated hamsters. At the same time when the BZD agonist was added together with the neuropeptide, ORX-2R mRNA levels moderately diminished (−48%) with respect to C in LHA and this effect resulted to be notably reduced (−174%) when compared to ORX-A-treated hamsters. Interestingly, while treatment with ORX-B alone was mainly responsible for moderate (36–45%) increases of ORX-2R mRNA densities in SON, CeA plus Lat, this trend instead proved to be notably (84%) enhanced in CeA when the neuropeptide was sequentially administered with NMDA (Fig. 7b). Even for these treatment sessions the transcriptional properties of the different neuronal fields resulted to be further enhanced as displayed by a numerically notably greater up-regulation for DG (365%) with respect to neuropeptide treatment alone while only a great entity was obtained for CeA (78%). Conversely, the addition of the BZD agonist strongly blocked the action of the ORXergic-related expression pattern especially in SON (−246%) and CeA (−190%) as shown by a still very great reduction in these limbic stations when compared to ORX-Btreated hamsters. Similarly, also the different treatments of CeA supplied comparable ORX-2R expression patterns for the above limbic areas. It was, however, ORX-B treatments that evoked notable effects in hypothalamic and AMY areas as indicated by a great (65%) and a moderate (32–40%) increases in BlA and in the hypothalamic areas (LHA, Pe and SON), respectively (Fig. 8b). This effect resulted to be of an even greater entity as demonstrated by a

Fig. 7. BlA treated with the two neuropeptides (ORX-A/B) alone or in combination with either the GABAergic (zolpidem) or glutamatergic (NMDA) agonists was next checked to evaluate their role on ORX-2R expression differences with respect to controls (represented as % change) in some limbic brain areas by using ANOVA plus a post hoc Newman–Keul’s test, *,a p < 0.05, **,b p < 0.01 and ***,c p < 0.001. For abbreviations check Fig. 1, while for asterisk and letter indications check Fig. 2.

notably great (90%) and moderate (50–60%) changes upon the addition of NMDA with respect to ORX-B-treated hamster aside Pe, which instead accounted for a great reduction (−70%). At the same time the moderate down-regulated levels of ORX-2R in DG (−43%) was greatly enhanced in the presence of the glutamatergic agonist (75%) with respect to C; an effect that instead proved to be notably up-regulated (290%) when compared to only ORX-B treated hamsters. Moreover, in presence of zolpidem a moderately down-regulation pattern was detected for BlA, which turned out to be of a notably great entity (−178%) with respect to ORX-B-treated hamsters. This trend was also detected for the two-hypothalamic nuclei, i.e. LHA (−273%) plus SON (−242%). ORX-A did not play a key role in this AMY station as shown by non-significant variations, aside a moderate down-regulation in LHA of hamsters infused with ORX-A alone while they became significant when administered in combination with either NMDA or zolpidem (Fig. 8a). This was the case for the great and moderate increases of ORX-2R with respect to C in BlA (72%) and DG (36%) of NMDA + ORX-A treated hamsters, at least for the latter brain station seems to behave in a parallel manner to feeding performances. Curiously, the moderate down-regulatory effects of ORX-A (−38%) in LHA, proved to be of a great (−62%) nature in the presence of the ␣1 GABAA agonist that in any case appeared to be of a moderate reduction with respect to the neuropeptide alone.

E. Avolio et al. / Behavioural Brain Research 234 (2012) 91–99

Fig. 8. CeA treated with the two neuropeptides (ORX-A/-B) alone or in combination with either the GABAergic (zolpidem) or glutamatergic (NMDA) agonists was next checked to evaluate their role on ORX-2R expression differences with respect to controls (represented as % change) in some limbic brain areas by using ANOVA plus a post hoc Newman–Keul’s test, *,a p < 0.05, **,b p < 0.01 and ***,c p < 0.001. For abbreviations check Fig. 1, while for asterisk and letter indications check Fig. 2.

4. Discussion In this first study, the contrasting feeding behaviors following the combined treatment of ORX-A/B with the agonist of either the major excitatory (glutamate) or inhibitory (GABAA R) neuroreceptor system of the two AMY nuclei (BlA and CeA) in the facultative hibernating hamster strongly underlie a differentiated role of ORXergic fibers cross-talking with amygdalar GABAergic/glutamatergic fibers. The hibernating rodent used in the present study is considered a useful model because the neuropeptide ORX-A/B, which are involved with the regulation of feeding and sleeping states considered critical for the different phases of hibernation [12]. As stated previously, hibernation is a normal physiological adaptation that leads animals to cease feeding habits along with a diminished body temperature in order to conserve energy stores. For this purpose our rodent model is retained advantageous over nonhibernating models in view of the greater brain plasticity properties occurring during some stages of hibernation and precisely that of the arousal state [47]. Indeed, induction of hyperphagic states by mostly BlA seems to occur in a conserved fashion to that evoked in non-hibernating rodents [53]. In the present work, variations of feeding behaviors deriving from tests conducted in constructed experimental cages considered appropriate for such paradigms allowed us to positively correlate eating/drinking rhythms to both neuropeptides, which tends to further support the importance of ORXergic-dependent appetite responses for the different vertebrate species [4]. The distinct feeding activities independent of

97

drinking bouts appear to be very tightly related to specific feeding stimulatory events and namely the various aspects of appetite, such as rate and size of eating episodes (gorging versus nibbling), energy levels of consumed foods plus variability in day-to-day food intake [15]. Even substantial neuronal activities of the different AMY nuclei seemed to account for increased food intake and consequently increased body weight like that of other rodents [45]. It appeared that AMY-dependent food restrictions were mostly correlated to the conservation of energy through a hyperphagicdependent reversal of nutritional deficit as recently shown by CeA influencing circulating leptin levels and consequently obesity states [55]. It should not be so surprising to observe that when the feeding capacity of hypothalamic and telencephalic ORXergic fibers (both ORX-A and B) was tested in the presence of the ␣1 GABAA R agonist, eating bouts were immediately reduced especially for ORX-A + zolpidem infusion within BlA [50]. In this case, BZD agonist accomplishes such a reduction very likely via sedative type of responses toward feeding motivational processes in a similar manner to that evoked by the ␤ GABAA R agonist muscimol on reward-dependent eating stimuli [51]. Interestingly, the other main AMY nucleus (CeA), known for its highly motivational behavioral drive displayed very few if any feeding stimulatory type of response for both neuropeptides. Failure of CeA to arouse eating stimulations could very well indicate that either this AMY region is not actively involved with such behaviors or that other factors may be inhibiting eating following the induction of AMY-dependent stressful conditions [5]. The latter condition tends to occur in a similar fashion to that of anxiogenic states in which by diminishing c-Fos immunoreactivity densities within CeA, aversive type of cues are removed [46] and thereby activating motivational signals required for eating events [40]. This condition fits very nicely with anxiolytic states of hibernators leading to the necessity of replenishing energy requests [48] and hence the reduction of aversive responsiveness causing animals to eat [35]. As a consequence, the clear-cut balance between GABAergic inhibitory events of CeA and the excitatory outputs of BlA toward feeding behavior support the homeostatic value of the posterodorsal amygdalar ORX-A system on hyperphagic responses through an anxiogenic type of mechanism [5]. It is worth noting that despite the notable effects of BlA ORXergic fibers on feeding behaviors in a very similar manner to that of hypothalamic and some telencephalic centers like the bed nucleus stria terminalis plus the ventral medulla along with the brainstem raphe pallidus, they do not show a similar regulatory capacity for drinking performances [56]. This was the case for the AMY nucleus treated with both neuropeptides not being able to modify drinking bouts, which supports the participation of other limbic areas such as the lateral/dorsomedial hypothalamic areas plus the subfornical organ and area postrema in satisfying drinking stimuli [34]. Interestingly, however, this did not seem to be the case for ORX-B fibers of CeA since they were responsible for frequent drinking bouts as shown by the very great consumption of water. Such a feature may very probably be linked to CeA being actively involved with antianxiety states [5] and so its antinociceptive effects on ORXergic fibers might rather, through the removal of the anxious state of the animal, favor explorative and drinking behaviors [18]. At the same time GABA seemed to reduce the efficacy of ORXergic influences and this tends to go along the same line of the pro-anxiety state being evoked through inhibitory GABAergic signals, which in turn cause animals to eat or drink less [35]. The inhibitory GABAergic effect of the BZD agonist on drinking is further supported by the blocking effects of zolpidem on hypothalamic (SON) and AMY (CeA and Lat) neuronal expression levels of ORX-2R. This seems to go in the same direction of the ␤GABAA R antagonist (bicuculline) inducing an activation of a thirst stimulus in ORX-treated animals, which

98

E. Avolio et al. / Behavioural Brain Research 234 (2012) 91–99

seems to be associated to an altered cellular transmembrane Cl− gradient during chronic hyperosmotic stress [24] and consequently satisfying physiological responses to body fluid imbalances such as water deprivation [19]. At the same time the inhibitory GABAergic signals reducing drinking stimuli tend to go in an opposite direction to that of the elevated ␣1 GABAA R mRNA levels inducing bingedrinking events [54]. Such an effect could very likely be due to the low concentration of zolpidem doses used in the present study not being able to establish the same conditions as the above elevated ␣1 mRNA levels, which is supported by greater concentrations of this BZD agonist leading to increased ORX-2Rs and consequently to greater licking and eventually drinking performances [11]. Moreover, a down-regulation of ORX-2R in LHA, SON, Lat along with BlA following ORX-A + ␣1 GABAA R agonist treatment trend appears to be in good agreement with hypothalamic-containing ␣1 GABAA Rs interfering not only with locomotor paradigms [57] but also with hyperphagic features [13]. As far as the excitatory circuit of the AMY complex is concerned, it appeared that the combined addition of NMDA + ORX-A in especially BlA induced mostly eating events as pointed out by increased eating frequencies and consequential body weight gain. The favorably greater NMDAergic-dependent eating activities should not be surprising since it goes in the same direction of other works in which rats treated with glutamate agonists displayed more time spent eating [3]. In addition, the preferential participation of the BlA glutamatergic fibers on feeding responses seems to strengthen its inclusion among the major limbic targets sites of excitatory-related eating responses such as the lateral hypothalamic and preoptic areas plus nAc [39]. It is possible that BlA NMDAergic fibers exert a positive role toward the acquisition of different environmental cues linked to feeding habits [14], via the activation of specific hypothalamic NMDA receptors promoting reward-dependent hyperphagic episodes in this hibernating rodent [2], while treatment with the NMDA receptor antagonist MK-801 seems to, instead, attenuate feeding stimuli [8]. Curiously, NMDA appeared to also influence drinking rhythms, even though of a lesser entity of that of feeding, as shown by the increased quantities of water consumed when this glutamatergic agonist was infused alone or together with both neuropeptides into mainly CeA. This specific activity tends to add to the already known ionotropic role of limbic area glutamate receptors favoring ORX-2R-dependent drinking behaviors [26] very likely through an NMDA-dependent increase of c-Fos activity [31], which by activating the hypothalamic vasopressin-ORXergic circuit induce elevated drinking rhythms [20]. Taken together, the findings of the present study indicate that BlA and CeA fibers by being heavily involved with food motivation stimuli are capable of controlling such a stimuli along with drinking behavior through distinct ON/OFF switching rhythms of AMY NMDAergic/GABAergic circuits. We are still at the beginning but indications arising from the present study may begin to unravel the role of AMY excitatory/inhibitory neuronal homeostatic conditions on ORX reward-dependent feeding and especially drinking behaviors [9] of hibernators. It is tempting to suggest that rodents undergoing exploratory type of behaviors may be induced to search for food so that they can replenish their energetic balance through AMY ORXergic-NMDAR/GABAergic circuits and this may prove to be an essential step for the induction of torpor state in hamsters [12]. In addition, altered drinking rhythms controlled by CeA seem to be consistent with GABAergic signals of this AMY site interfering with environmentally-related thirst stimuli through the reduction of anxiety-induced aversive cues [27] and thereby causing ORX fibers to activate hypothalamic neurons within SON that control body fluid balance. As a consequence, this type of mechanism plus implications deriving from recently reported ghrelin-related ORXgenic effects accounting for increased feeding plus fluid/water consumption [17] may suggest new therapeutic values of ORX

fibers on drinking and feeding disorders such as psychotic polydipsia, dehydrated anorexia and obesity [29]. Acknowledgments We gratefully thank the Italian University Research Ministry (MIUR), European Commission and Calabria Region for having cofinanced this study. References [1] Ahn S, Phillips AG. Modulation by central and basolateral amygdalar nuclei of dopaminergic correlates of feeding to satiety in the rat nucleus accumbens and medial prefrontal cortex. Journal of Neuroscience 2002;22:10958–65. [2] Alò R, Avolio E, Carelli A, Facciolo RM, Canonaco M. Amygdalar glutamatergic neuronal systems play a key role on the hibernating state of hamsters. BMC Neuroscience 2011;12:10–21. [3] Andrzejewski ME, Sadeghian K, Kelley AE. Central amygdalar and dorsal striatal NMDA receptor involvement in instrumental learning and spontaneous behavior. Behavioral Neuroscience 2004;118:715–29. [4] Anukulkitch C, Rao A, Pereira A, McEwan J, Clarke LJ. Expression of genes for appetite-regulating peptides in the hypothalamus of genetically selected lean and fat sheep. Neuroendocrinology 2010;91:223–38. [5] Avolio E, Alò R, Carelli A, Canonaco M. Amygdalar orexinergic–GABAergic interactions regulate anxiety behaviors of the Syrian golden hamster. Behavioural Brain Research 2011;218:288–95. [6] Baldo BA, Kelley AE. Discrete neurochemical coding of distinguishable motivational processes: insights from nucleus accumbens control of feeding. Psychopharmacology 2007;191:439–59. [7] Barnard EA, Skolnich P, Olsen RW, Mohler H, Sieghart W, Biggio G, Braestrup C, Bateson AN, Langer SZ. Subtypes of gamma-aminobutyric acid A receptors: classification on the basis of subunit structure and receptor function. Pharmacological Reviews 1998;50:291–313. [8] Bellinger LL, Bernardis LL. The dorsomedial hypothalamic nucleus and its role on ingestive behavior and body weight regulation: lessons learned from lesioning studies. Physiology and Behavior 2002;76:431–42. [9] Berridge KC, Ho CY, Richard JM, Di Feliceantonio AG. The tempted brain eats: pleasure and desire circuits in obesity and eating disorders. Brain Research 2010;1350:43–64. [10] Chemelli RM, Willie JT, Sinton CM, Elmquist JK, Scammell T, Lee CE, Richardson JA, Williams SC, Xiong Y, Kisanuki Y, Fitch TE, Nakazato M, Hammer RE, Saper CB, Yanagisawa M. Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell 1999;98:437–51. [11] Cooper SJ, Higgs S. Benzodiazepine effects on licking responses for sodium chloride solutions in water-deprived male rats. Physiology and Behavior 2005;85:252–8. [12] Drew KL, Buck CL, Barnes M, Christian SL, Rasley BT, Harris MB. Central nervous system regulation of mammalian hibernation: implications for metabolic suppression and ischemia tolerance. Journal of Neurochemistry 2007;102:1713–26. [13] Funato H, Tsai AL, Willie JT, Kisanuki Y, Williams SC, Sakurai T, Yanagisawa M. Enhanced orexin receptor-2 signaling prevents diet-induced obesity and improves leptin sensitivity. Cell Metabolism 2009;9:64–76. [14] Galarce EM, McDannald MA, Holland PC. The basolateral amygdala mediates the effects of cues associated with meal interruption on feeding behavior. Brain Research 2010;1350:112–22. [15] Gargiulo PA. Thyrotropin releasing hormone injected into the nucleus accumbens septi selectively increases face grooming in rats. Brazilian Journal of Medical and Biological Research 1996;29:805–10. [16] Halford JC, Wanninayake SCD, Blundell JE. Behavioral satiety sequence (BSS) for the diagnosis of drug action on food intake. Pharmacology Biochemistry and Behavior 1998;61:159–68. [17] Hashimoto H, Ueta Y. Central effects of ghrelin, an unique peptide on appetite and fluid/water drinking behavior. Current Protein & Peptide Science 2011;12:280–7. [18] Ho YC, Lee HJ, Tung LW, Liao YY, Fu SY, Teng SF, Liao HY, Mackie K, Chiou LC. Activation of orexin 1 receptors in the periacqueductal gray of male rats leads to antinociception via retrograde endocannabinoid (2-arachidonoylglycerol)induced disinhibition. Journal of Neuroscience 2011;31:14600–10. [19] Houston AJ, Wong JC, Ebenezer IS. Effects of subcutaneous administration of the gamma-aminobutyric acid(A) receptor agonist muscimol on water intake in water-deprived rats. Physiology and Behavior 2002;77:445–50. [20] John J, Kumar VM. Effect of NMDA lesion of the medial preoptic neurons on sleep and other functions. Sleep 1998;21:587–98. [21] Kelley AE. Ventral striatal control of appetitive motivation: role in ingestive behavior and reward-related learning. Neuroscience and Biobehavioral Reviews 2004;27:765–76. [22] Kelley AE, Baldo BA, Pratt WE, Will MJ. Corticostriatal-hyphotalamic circuitry and food motivation: integraton of energy, action and reward. Physiology and Behavior 2005;86:773–95. [23] Kia HK, Yen G, Krebs CJ, Pfaff DW. Co-localization of estrogen receptor alpha and NMDA-2D mRNAs in amygdaloid and hypothalamic nuclei of the mouse brain. Molecular Brain Research 2002;104:47–54.

E. Avolio et al. / Behavioural Brain Research 234 (2012) 91–99 [24] Kim JS, Kim WB, Kim YB, Lee Y, Kim YS, Shen FY, Lee SW, Park D, Choi HJ, Hur J, Park JJ, Han HC, Colwell CS, Cho YW, Kim YI. Chronic hyperosmotic stress converts GABAergic inhibition into excitation in vasopressin and oxytocin neurons in the rate. Journal of Neuroscience 2011;31:13312–22. [25] LeDoux J. The amygdala. Current Biology 2007;17:R868–74. [26] Li FW, Deurveilher S, Semba K. Behavioural and neuronal activation after microinjections of AMPA and NMDA into the perfornical lateral hypothalamus in rats. Behavioural Brain Research 2011;224:376–86. [27] Lyons AM, Thiele TE. Neuropeptide Y conjugated to saporin alters anxiety-like behavior when injected into the central nucleus of the amygdala or basomedial hypothalamus in BALB/cJ mice. Peptides 2010;31:2193–9. [28] Marcus JN, Aschkenasi CJ, Lee CE, Chemelli RM, Saper CB, Yanagisawa M, et al. Differential expression of orexin receptors 1 and 2 in the rat brain. Journal of Comparative Neurology 2001;435:6–25. [29] Milella MS, Passarelli F, De Carolis L, Schepisi C, Nativio P, Scaccianoce S, Nencini P. Opposite roles of dopamine and orexin in quinpirole-induced excessive drinking: a rat model of psychotic polydipsia. Psychopharmacology 2010;211:355–66. [30] Morin LP, Wood RI. A stereotaxic atlas of the golden hamster brain. San Diego, CA: Academic Press; 2000. [31] Mota-Ortiz SR, Sukikara MH, Bittencourt JC, Baldo MV, Elias CF, Felicio LF, Canteras NS. The periacqueductal gray as a critical site to mediate reward seeking during predatory hunting. Behavioural Brain Research 2012;226:32–40. [32] Nakashima M, Uemura M, Yasui K, Ozaki HS, Tabata S, Taen A. An anterograde and retrograde tract-tracing study on the projections from the thalamic gustatory area in the rat: distribution of neurons projecting to the insular cortex and amygdaloid complex. Neuroscience Research 2000;36:297–309. [33] Nambu T, Sakurai T, Mizukami K, Hosoya Y, Yanagisawa M, Goto K. Distribution of orexin neurons in the adult rat brain. Brain Research 1999;827:243–60. [34] Ono K, Kai A, Honda E, Inenaga K. Hypocretin-1/orexin-A activates subfornical organ neurons of rats. Neuroreport 2008;19:69–73. [35] Olszewski PK, Grace MK, Fard SS, Le Grevès M, Klockars A, Schiöth HB, Levine AS. Central nociceptin/orphanin FQ system elevates food consumption by both increasing energy intake and aversive responsiveness. American Journal of Physiology - Regulatory, Integrative and Comparative Physiology 2010;299:653–63. [36] Olsen RW, Sieghart W. GABAA receptors: subtypes provide diversity of function and pharmacology. Neuropharmacology 2009;56:141–8. [37] Ottoni EB. EthoLog 2.2: a tool for the transcription and timing of behavior observation sessions. Behavior Research Methods, Instruments, & Computers 2000;32:446–59. [38] Patrick L. Eating disorders: a review of literature with emphasis on medical complications and clinical nutrition. Alternative Medicine Review 2002;7:84–202. [39] Petrovich GD, Holland PC, Gallagher M. Amygdalar and prefrontal pathways to the lateral hypothalamus are activated by a learned cue that stimulates eating. Journal of Neuroscience 2005;25:8295–302. [40] Petrovich GD, Ross CA, Mody P, Holland PC, Gallagher M. Central but basolateral, amygdala is critical for control of feeding by aversive learned cues. Journal of Neuroscience 2009;29:15205–12. [41] Reynolds SM, Berridge KC. Glutamate motivational ensembles in nucleus accumbens: rostrocaudal shell gradients of fear and feeding. European Journal of Neuroscience 2003;17:2187–200.

99

[42] Sah P, Faber ES, Lopez DA, Power J. The amygdaloid complex: anatomy and physiology. Physiological Reviews 2003;83:803–34. [43] Sakurai T, Amemiya A, Ishii M, Matsuzaki I, Chemelli RM, Tanaka H, Williams SC, Richardson JA, Kozlowski GP, Wilson S, Arch JR, Buckingham RE, Haynes AC, Carr SA, Annan RS, McNulty DE, Liu WS, Terrett JA, Elsourbagy NA, Bergsma DJ, Yanagisawa M. Orexins and orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding and behavior. Cell 1998;92:573–85. [44] Sayander V, Le Doux JE, Pitkanen A. Interamygdaloid projections of the basal and accessory basal nuclei of the rat amygdaloid complex. Neuroscience 1997;76:725–35. [45] Seamans JK, Phillips AG. Selective memory impairments produced by transient lidocaine-induced lesions of the nucleus accumbens in rats. Behavioral Neuroscience 1994;108:456–68. [46] Shaw D, Norwood K, Leslie JC. Chlordiazepoxide and lavener oil alter unconditioned-induced c-fos expression in the rat brain. Behavioural Brain Research 2011;224:1–7. [47] Stenzel-Poore MP, Stevens SL, Xiong Z, Lessov NS, Harrington CA, Mori M, Meller R, Rosenzweig HL, Tobar E, Shaw TE, Chu X, Simon RP. Effect of ischemic preconditioning on genomic response to cerebral ischaemia: similarity to neuroprotective strategies in hibernation and hypoxia- tolerant states. Lancet 2003;362:1028–37. [48] Storey KB, Storey JM. Putting life on ‘pause’ – molecular regulation of hypometabolism. Journal of Experimental Biology 2007;210:1700–14. [49] Swanson LW. The amygdala and its place in the cerebral hemisphere. Annals of the New York Academy of Sciences 2003;985:174–84. [50] Thorpe AJ, Cleary JP, Levine AS, Kotz CM. Centrally administered orexin A increases motivation for sweet pellets in rats. Psychopharmacology 2005;182:75–83. [51] Wellman LL, Gale K, Malkova L. GABAA-mediated inhibition of basolateral amygdala blocks reward devaluation in macaques. Journal of Neuroscience 2005;25:4577–86. [52] Winrow CJ, Tanis KQ, Reiss DR, Rigby AM, Uslaner JM, Uebele VN, Doran SM, Fox SV, Garson SL, Gotter AL, Levine DM, Roecker AJ, Coleman PJ, Koblan KS, Renger JJ. Orexin receptor antagonism prevent transcriptional and behavioral plasticity resulting from stimulant exposure. Neuropharmacology 2010;58: 185–93. [53] Xu J, Kirigiti MA, Groye KL, Smith MS. Regulation of food intake and gonadotropin-releasing hormone/luteinizing hormone during lactation: role of insulin and leptin. Endocrinology 2009;150:4231–40. [54] Yang X, Yan J, Lu B, Zhao X, Lei Q, Yang D, Chen K, Zhao S, Zhu G. Fos expression and hormone changes following electrical stimulation of the posterodorsal amygdale and the effects on food intake in conscious female rats. Brain Research 2009;1273:83–91. [55] Yang AR, Liu J, Yi HS, Warnock KT, Wang M, June HL, Puche AC, Elnabawi A, Sieghart W, Aurelian L, June HL. Binge drinking: in search of its molecular target via the GABA(A) receptor. Frontiers in Neuroscience 2011;5:123. [56] Yu L, Zhang XY, Zhang J, Zhu JN, Wang JJ. Orexins excite neurons of the rat cerebellar nucleus interpositus via orexin 2 receptors in vitro. Cerebellum 2010;9:88–95. [57] Zheng H, Patterson LM, Berthoud HR. Orexin-A projections to the caudal medulla and orexin-induced c-Fos expression, food intake and autonomic function. Journal of Comparative Neurology 2005;485:127–42.