The Orexigenic Effect of Orexin-A Revisited: Dependence of an Intact Growth Hormone Axis

Fifteen years ago orexins were identified as central regulators of energy homeostasis. Since then, that concept has evolved considerably and orexins are currently considered, besides orexigenic neuropeptides, key modulators of sleep-wake cycle and neuroendocrine function. Little is known, however, about the effect of the neuroendocrine milieu on orexins' effects on energy balance. We therefore investigated whether hypothalamic-pituitary axes have a role in the central orexigenic action of orexin A (OX-A) by centrally injecting hypophysectomized, adrenalectomized, gonadectomized (male and female), hypothyroid, and GH-deficient dwarf rats with OX-A. Our data showed that the orexigenic effect of OX-A is fully maintained in adrenalectomized and gonadectomized (females and males) rats, slightly reduced in hypothyroid rats, and totally abolished in hypophysectomized and dwarf rats when compared with their respective vehicle-treated controls. Of note, loss of the OX-A effect on feeding was associated with a blunted OX-A-induced increase in the expression of either neuropeptide Y or its putative regulator, the transcription factor cAMP response-element binding protein, as well as its phosphorylated form, in the arcuate nucleus of the hypothalamus of hypophysectomized and dwarf rats. Overall, this evidence suggests that the orexigenic action of OX-A depends on an intact GH axis and that this neuroendocrine feedback loop may be of interest in the understanding of orexins action on energy balance and GH deficiency.

Orexins/hypocretins (orexin-A/hypocretin 1: OX-A/Hcrt1 and orexin-B/hypocretin 2: OX-B/Hcrt2) are hypothalamic neuropeptides, mainly expressed in the lateral hypothalamus (LHA), that are produced from a common precursor called prepro-OX or prepro-Hcrt (1, 2). Although orexins were originally identified as regulators of food intake and energy balance (1, 3–7) and later as key modulators of sleep-wake cycle and arousal (8–10), multiple lines of evidence have demonstrated a role of orexins in the control of practically all the endocrine axes (5, 11–17). Evidence supporting the latter is as follows. OX-A administration activates the hypothalamic-pituitary-adrenal axis by acting at the CRH neurons in the paraventricular nucleus of the hypothalamus, as well as by increasing circulating levels of ACTH and corticosterone (CORT) (16). Orexins are also involved in the modulation of the hypothalamic-pituitary-gonadal axis by controlling the secretion of LH from gonadotropes (18) and directly acting on follicle development in the ovary in females (19), as well as controlling male spermatogenesis and testosterone production in the testicle (17, 20). The hypothalamic-pituitary-thyroid (HPT) is also regulated by orexins mainly at hypothalamic level (21, 22). Finally, OX-A has been also shown to diminish endogenous plasma GH levels (23) by inhibiting the pulsatile GH secretion in the rat through a mechanism involving somatostatin neurons in the periventricular nucleus of the hypothalamus and GHRH neurons in the paraventricular nucleus of the hypothalamus (14, 24).

However, despite this evidence the possible influence that neuroendocrine status may have on orexin effects on energy balance remains unknown. This question is of relevance because alterations in the hypothalamic-pituitary function exert a significant impact on energy metabolism. In fact, it is well established that total loss of pituitary function (25–27) or deficiency of adrenal (28, 29), gonadal (30), thyroid (12, 31, 32), or GH (ie, dwarf rats) (33–36) axes promote alterations in feeding patterns and whole-body energy metabolism. On the basis of that evidence, the aim of this study has been to investigate whether OX-A-induced feeding is modulated by changes in the neuroendocrine system. For that purpose, we studied the effect of central OX-A administration on hypophysectomized (HPX), adrenalectomized (ADX), gonadectomized (GNX, females and males), hypothyroid, and GH-deficient dwarf rats. Our data showed that OX-A elicits food intake in ADX, GNX (females and males), and hypothyroid rats but it fails to do so in HPX and dwarf GH-deficient rats. Altogether these data suggest that the orexigenic action of OX-A requires an intact GH axis.

Materials and Methods

Animals

We used 4 different rat models: 1) male and female Sprague Dawley rats (250–300 g; Animalario General USC; Santiago de Compostela, Spain); 2) male HPX Sprague Dawley rats (250–300 g; Charles River; L'Arbresle, France); 3) male dwarf (HsdOla: dw-4) Lewis rats (175–200 g; Harlam Ibérica; Barcelona, Spain); and 4) male Lewis rats (175–200 g; Harlam Ibérica). All animals were housed on a 12-hour light (8:00 am to 8:00 pm), 12-hour dark cycle, in a temperature- (22–24°C) and humidity-controlled room. The animals were allowed free access to standard laboratory pellets of rat chow (Scientific Animal Food & Engineering [SAFE]; Nantes, France) and tap water. The experiments were performed in agreement with the International Law on Animal Experimentation and were approved by the USC Local Ethical Committee and the Ministry of Science and Innovation of Spain (Project ID PI12/01814).

Surgical procedures and validation of adrenalectomy and gonadectomy

In order to analyze the effect of gonadal hormones on OX-A orexigenic effect, male and female Sprague Dawley rats were bilaterally orchidectomized (ORX), ovariectomized (OVX), or sham-operated as described previously (30). Absence of gonadal function was confirmed by increased LH serum levels measured using a double-antibody method and RIA kits supplied by the National Institutes of Health (Dr. A. F. Parlow; National Institute of Diabetes and Digestive and Kidney Diseases National Hormone and Peptide Program, Torrance, California) as reported elsewhere (30). Central treatment with OX-A was carried 3 weeks after surgery.

Bilateral adrenalectomy of rats was performed as previously reported (37, 38), whereas sham-adrenalectomy was performed exactly the same way without removing the adrenal glands. Total lack of adrenal function was confirmed by showing complete absence of serum CORT immunoreactivity (Linco Research Assay Services; St. Charles, Missouri) (39) at the end of the study. The drinking water of adrenalectomized rats was replaced with a 5% glucose and 0.9% sodium chloride solution instead to prevent hypoaldosteronism-induced hyponatremia. Central treatment with OX-A was performed 3 weeks after surgery.

Induction and validation of hypothyroidism

Hypothyroidism was induced as previously described by administration of 0.1% aminotriazole (Sigma, St Louis, Missouri) in drinking water for a period of 3 weeks (12, 31). Plasma levels of T₃ and T₄ were measured using rat ELISA kits (Crystal Chem Inc; Downers Grove Illinois) (12, 31, 32).

Validation of hypophysectomized and dwarf rats

Male HPX rats form Charles River and male dwarf rats from Harlam Ibérica exhibited a clear phenotype in terms of reduced body length and body weight. Furthermore, IGF-1 levels were assayed using a Quantikine ELISA, mouse/rat IGF-1 kit, (R&D Systems, Minneapolis, Minnesota).

Implantation of intracerebroventricular cannulae and OX-A treatment

Chronic intracerebroventricular (ICV) cannulae were implanted under ip ketamine-xylazine anesthesia (200 μL/100 g body weight; 42.5% ketamine [50 mg/mL; Ketalar, Parke-Davis, Morris Plains, New Jersey], 20% [2 mg/mL; Rompum, Bayer, Leverkusen, Germany] and 37.5% saline), as described previously, and correct positioning in the lateral ventricle was confirmed by postmortem histologic examination (3, 4, 12, 26, 27, 30–32, 39–49). The animals were caged individually and used for experimentation 4 days later. During this postoperative recovery period the rats became accustomed to the handling procedure under nonstressful conditions. Rats received either a single administration of OX-A (Bachem; Bubendorf, Switzerland) (3 nmol [=10 μg] per rat dissolved in 5 μL of saline) or vehicle (control rats). The selection of this dose was based on previous reports (1, 4, 14, 24). HPX rats were treated 10–14 days (14 days at latest) after they were hypophysectomized. In the case of hypothyroid rats, AMT was given continuously in drinking water during the cannulae placement and OX-A treatment. Animals were treated at 9:00 am (1 hour after the light cycle had commenced), when they were satiated. Rats were killed by cervical dislocation. For the expression analyses, animals were killed 2 hours after OX-A administration. From each animal, the arcuate nucleus of the hypothalamus (ARC) (for Western blotting) was immediately homogenized on ice to preserve phosphorylated protein levels or the whole brain (for in situ hybridization) were dissected, and stored at −80°C until further processing. Dissection of the ARC was performed by micropunches under the microscope, as previously shown (31, 32, 47).

Western blotting

Hypothalamic and ARC protein lysates were subjected to SDS-PAGE, electrotransferred on a polyvinylidene difluoride membrane, and probed with the following antibodies: CREB (cAMP response element-binding protein) and pCREB (phosphorlated CREB)-Ser¹³³ (Santa Cruz Biotechnology, Inc.; Santa Cruz, California) and β-actin (Sigma) as previously described (26; 27; 30–32; 41; 42; 44–50).

In situ hybridization

Coronal brain sections (16 μm) were probed with specific oligonucleotides for neuropeptide Y (NPY) (GenBank Accession no.: M20373; 5′-AGA TGA GAT GTG GGG GGA AAC TAG GAA AAG TCA GGA GAG CAA GTT TCA TT-3′) and prepro-OX (GenBank Accession no.: NM_013179; 5′-TTC GTA GAG ACG GCA GGA ACA CGT CTT CTG GCG ACA-3′) as previously published (3, 4, 12, 27, 31, 32, 39, 41,42,44–47, 49, 50).

Statistical analysis

Food intake data (mean ± SEM) are represented as cumulative feeding 2, 4, and 6 hours after the treatment. mRNA and protein data were expressed as mean ± SEM in relation (%) to control (vehicle-treated or normal) rats. Statistic significance was determined by Student's t test when 2 groups were compared or ANOVA and post hoc two-tailed Bonferroni test when more than 2 groups were compared. P < .05 was considered significant. The number of animals used in each experimental setting is specified in each figure.

Results

OX-A-induced feeding and NPY expression in the ARC of normal male rats

ICV administration of OX-A significantly increased food intake when administered to satiated rats. This orexigenic effect was transient, being significant 2, 4, and 6 hours after the treatment and disappearing 12 and 24 hours after the administration of the OX-A (Figure 1A). Using in situ hybridization analysis, a significant increase of NPY mRNA content was observed in the ARC after the treatment with OX-A (Figure 1, B and C).

Figure 1.

Effects of Central OX-A Administration on Normal Male Rats. Cumulative food intake (panel A), in situ hybridization autoradiographic images (panel B), and NPY mRNA levels (panel C) in the ARC of normal male rats ICV treated with OX-A. 3V, third ventricle; DMH, dorsomedial nucleus of the hypothalamus. ***, P < .001 vs vehicle; 9–12 rats per group were used.

OX-A failed to induce feeding and NPY expression in the ARC of HPX male rats

HPX male rats showed significantly reduced serum IGF-1 levels (Figure 2A), confirming the efficiency of the hypophysectomy procedure. ICV administration of OX-A significantly increased food intake when administered to satiated normal intact rats at any evaluated time (2, 4, and 6 hours). On the contrary, such orexigenic effect was totally absent when OX-A was centrally administered to HPX rats (Figure 2B). Of note, lack of the orexigenic action was associated with unaltered NPY mRNA expression in the ARC of HPX rats following OX-A injection (Figure 2, C and D). It is important to highlight that NPY mRNA levels were constitutively decreased in the ARC of HPX rats when compared with their normal intact controls (Supplemental Figure 1,A and B, published on The Endocrine Society's Journals Online web site at http://endo.endojournals.org). Our data showed no alterations in the prepro-OX mRNA in the LHA of HPX rats, indicating that decreased NPY mRNA levels are not due to impaired endogenous orexinergic tone (Figure 2, E and F).

Figure 2.

Effects of Central OX-A Administration on HPX Male Rats. IGF-1 serum levels (panel A), cumulative food intake (panel B), in situ hybridization autoradiographic images (panel C), and NPY mRNA levels (panel D) in the ARC of normal and HPX male rats ICV treated with OX-A. In situ hybridization autoradiographic images (panel E) and prepro-OX mRNA levels (panel F) in the LHA of normal and HPX male rats. 3V, third ventricle; DMH, dorsomedial nucleus of the hypothalamus. *, P < .05 vs normal vehicle; **, P < .001 vs normal vehicle; 7–16 rats per group were used.

OX-A-induced feeding in gonadectomized female and male rats

Having shown that OX-A fails to induce feeding in a model of total lack of pituitary function in HPX rats, we wanted to evaluate the contribution of each neuroendocrine axis to that lack of response. OVX female and ORX male rats showed the expected increase in LH serum levels 3 weeks after the ovaries or the testes, respectively, were removed (Figure 3, A and C), confirming the efficiency of the gonadectomy procedures. ICV administration of OX-A significantly induced food intake in both OVX female and ORX male rats to a degree similar to that observed in their respective sham-operated controls (Figure 3, B and D). Prepro-OX mRNA levels in the LHA of gonadectomized rats are similar to sham-operated controls (51).

Figure 3.

Effects of Central OX-A Administration on GNX Female and Male Rats. Serum LH levels in female OVX (panel A) and male ORX rats (panel C). Cumulative food intake in sham-operated and female OVX (panel B) and male ORX rats (panel D) ICV treated with OX-A. ***, P < .001 vs sham vehicle; ##, P < .01 vs OVX vehicle; ###. P < .001 vs OVX vehicle or ORX vehicle; 8–9 rats per group were used.

OX-A-induced feeding in ADX male rats

ADX male rats showed undetectable CORT serum levels (Figure 4A), confirming the efficiency of the adrenalectomy procedure. ICV administration of OX-A significantly induced feeding in ADX rats in a magnitude similar to that shown in sham-operated controls (Figure 4B). Prepro-OX mRNA levels in the LHA of ADX rats were significantly reduced when compared with sham controls (sham-operated: 100 ± 8.9 vs ADX: 66.1 ± 6.7; P < .01).

Figure 4.

Effects of Central OX-A Administration on ADX and Hypothyroid Male Rats. Serum CORT levels in male ADX rats (panel A). Cumulative food intake (panel B) of sham-operated and ADX male rats ICV treated with OX-A. Serum T₃ (panel C) and T₄ levels (panel D) and cumulative food intake (panel E) in male euthyroid and hypothyroid rats ICV treated with OX-A. ***, P < .001 vs sham vehicle or euthyroid vehicle; ##, P < .01 vs ADX vehicle or hypothyroid vehicle; ###, P < .001 vs. ADX vehicle or hypothyroid vehicle. !!, P < .01 euthyroid OX-A vs hypothyroid OX-A; 8–10 rats per group were used. ND, not detected.

OX-A-induced feeding in hypothyroid male rats

Male rats receiving AMT in drinking water showed a significant decrease in the circulating levels of T₃ and T₄ (Figure 4, C and D), verifying their hypothyroid status. Interestingly, when centrally treated with OX-A, hypothyroid rats showed an increase in food intake but to a lesser degree than euthyroid controls (Figure 4E). Prepro-OX mRNA levels in the LHA of hypothyroid rats did not show any difference when compared with euthyroid rats (12).

OX-A failed to induce feeding and NPY expression in the ARC of dwarf male rats

Next, we investigated the orexigenic effect of central OX-A administration in GH-deficient dwarf rats (14, 52). Dwarf male rats showed significantly reduced serum IGF-1 levels (Figure 5A), validating their phenotype. Our data showed that OX-A lacked its orexigenic effect in our animal model of GH deficiency (Figure 5B). This lack of effect was associated to an impaired response in NPY mRNA expression, which was unaltered after OX-A treatment in dwarf rats (Figure 5. C and D). Of note, NPY mRNA levels were constitutively decreased in the ARC of dwarf rats when compared with their controls (Supplemental Figure 1, C and D) (27). To eliminate impaired endogenous orexinergic tone as at the root of this alteration, we examined the expression of prepro-OX in their hypothalami. Our data showed no alterations in the prepro-OX mRNA in the LHA of dwarf rats (Figure 5, E and F).

Figure 5.

Effects of Central OX-A Administration on GH-Deficient Dwarf Male Rats. IGF-1 serum levels (panel A), cumulative food intake (panel B), in situ hybridization autoradiographic images (panel C), and NPY mRNA levels (panel D) in the ARC of normal and GH-deficient dwarf male rats ICV treated with OX-A. In situ hybridization autoradiographic images (panel E) and prepro-OX mRNA levels (panel F) in the LHA of normal and GH-deficient dwarf male rats. 3V, third ventricle; DMH, dorsomedial nucleus of the hypothalamus. *, P < .05 vs normal vehicle; 7–15 rats per group were used.

OX-A increases CREB and pCREB protein expression in the ARC of normal but not in HPX and dwarf rats

Having shown that OX-A induced feeding and NPY mRNA expression in normal rats but not in HPX and dwarf rats, we investigated the effect of this treatment on the expression levels of CREB and its phosphorylated form, pCREB, a key transcription factor regulating Npy gene transcription. Our results showed that administration of OX-A induced increases in the protein content of both CREB and pCREB in the ARC of normal rats (Figure 6, A and B) but not in HPX (Figure 6, C and D) and dwarf rats (Figure 6, E and F). These data suggest that the OX-A effect on Npy gene expression, and subsequently on feeding, requires an intact GH axis.

Figure 6.

Effects of Central OX-A Administration on CREB and pCREB Protein Content in the ARC of Normal, HPX, and GH-Deficient Dwarf Rats. Western blot autoradiographic images and protein levels of pCREB and CREB in the ARC of normal (panels A and B) HPX (panels C and D) and GH-deficient dwarf rats (panels E and F). *, P < .05 vs vehicle; 6–8 rats per group were used.

Discussion

In this study, we first demonstrate that the orexigenic effect of OX-A requires an intact somatotropic axis and that action elicits increased levels of NPY mRNA expression in the ARC, through modulation of the transcription factor CREB and its phosphorylated isoform pCREB.

Over the last 15 years, compelling evidence has demonstrated that, among other physiologic processes, orexins play a major role in the modulation of energy balance (1, 3–5, 7, 53) and neuroendocrine system (5, 11–13, 15–17,54), with particular relevant actions on the GH axis (14, 15, 24). Given the intimate interrelationship and the precise feedback regulation that all these neuroendocrine axes display, we speculated that OX-A actions on feeding might be modulated by pituitary function. To assess this hypothesis, we started our experiments by using HPX rats, a model in which all the endocrine axes, except prolactin, are blunted as a consequence of the total loss of pituitary function (14, 34). Our data showed that OX-A does not stimulate either feeding or Npy or CREB/pCREB expression in the ARC of HPX rats, indicating that a normal pituitary function is needed for orexin to accomplish its orexigenic action. Dissection of the individual contribution of each pituitary-endocrine axis to OX-A-induced food intake revealed that a deficient adrenocorticotropic or gonadotropic axis does not impair the central response to OX-A. Notably, when OX-A was given to hypothyroid rats, the orexigenic effect was maintained but to a lesser degree than that observed in euthyroid rats. One interesting feature of hypothyroid rats is that, in addition to having low thyroid hormone levels, they show diminished circulating GH levels and GH content in the pituitary (55). This evidence, combined with the total lack of response to central OX-A observed in HPX rats, might indicate that the GH status may mediate the orexigenic response to OX-A. To check such a possibility, we centrally treated GH-deficient dwarf rats with OX-A; this GH-null strain arose in an inbred colony by a spontaneous mutation that affects regulatory sequences, because the genomic structure of the GH gene is apparently unaltered (52, 56–58). Our data showed that OX-A failed to stimulate food intake in dwarf rats, indicating that an intact GH axis is indispensable for OX-A to induce appetite. Our results also showed that this action was associated with a lack of effect of OX-A on Npy gene expression, as well as its putative transcription factor CREB/pCREB in the ARC. This evidence is in agreement with a recent study demonstrating that increased orexin signaling stimulates pCREB expression in mouse hypothalamic cell lines (59). This action also provides a putative molecular mechanism for the impaired response to OX-A in dwarf (and also in HPX) rats, because the orexin-NPY neuronal circuit (60) plays a fundamental role in orexin actions on feeding. In fact, selective inhibition of NPY Y1 and Y5 receptors blocks feeding stimulation induced by OX-A (53, 61), a phenomenon that is also observed in HPX and dwarf rats, which display constitutively decreased levels of NPY expression in the ARC (27), despite showing higher levels of CREB and pCREB. These observations are consistent with the results previously described in other studies, including ours (27, 57, 62–64) and also with the fact that most NPY neurons in the ARC express GH receptor mRNA, and the binding of GH to its receptor regulates those neurons and induces the expression of NPY mRNA (62, 65–67). The strong dependence of OX-A on endogenous GH secretory status for response to food intake appears to be quite specific for the orexin system. Previous studies have shown that although ghrelin orexigenic effect is also impaired in HPX rats, in this case the lack of response to ghrelin was due to impaired function of the pituitary-adrenal and/or thyroid axes (25).

The physiologic relevance of the OX-A-somatotropic axis interaction in the modulation of feeding is intriguing. First, orexins may act as a signal linking metabolic and nutritional status with the somatotropic axis. GH plays a major role in the regulation of metabolism, and impairment of the GH axis induces major changes in substrate mobilization, mainly glucose and lipids, in the liver and white adipose tissue (33–36;68). In this regard, dwarf rats, GH-deficient patients, and GH receptor-null (GH receptor-knockout) mice (26, 27, 33, 35, 69) all display impaired lipid metabolism and/or decreased insulin secretion. Thus it is possible that metabolic alterations, alongside distorted hormonal status, in those models may alter the short-term appetite response induced by OX-A. In this regard, recent evidence from our group has demonstrated that lack of GH impairs the normal physiologic responses of hypothalamic energy sensors, such as fatty acid synthase and AMP-activated protein kinase, and neuropeptides, such as NPY and agouti-related protein, induced by fasting or ghrelin administration. In fact, even in an ad libitum fed state, dwarf rats show decreased fatty acid synthase expression and increased AMP-activated protein kinase activation, a situation that mimics starvation yet does not induce a counterregulatory hyperphagic response in this rat model (27). Overall, this evidence suggests that a lack of GH in dwarf rats impairs the normal adaptive mechanisms to energy sensors and their neuropeptide effectors, putatively including orexins, and their short-term actions on feeding. From a teleologic point of view, this possibility seems likely because the induction of a short but massive orexigenic reaction would not be functionally relevant in a context where the lack of GH (plus low insulin levels) would not provide an appropriate hormone milieu for the rapid mobilization of the ingested substrates to allow an efficient storage of calories. In keeping with this, dwarf rats display an overall lean phenotype despite reduced adiposity and hypoleptinemia (70). Finally, whether these findings are exerted by GH itself and/or mediated by the associated IGF-1 deficiency is yet unclear. IGF-1 receptors are present in the hypothalamus, and a permissive role for brain IGF-1 receptors in compensatory hyperphagia induced by anesthesia and surgical manipulation has been reported (71). By contrast, IGF-1 infusion into the brain did not reduce food intake after 2 days in normal male rats (72). To our knowledge no data are available regarding the expression of prepro-orexin in appropriate animal models of pure IGF-1 deficiency; thus, further studies are needed.

In summary, our study shows that an intact GH axis is required for the manifestation of OX-A actions on food intake, which are conducted through increased CREB phosphorylation and Npy gene expression in the ARC. This neuroendocrine feedback loop may be of interest in the understanding of orexin's action on energy balance and may open new therapeutic possibilities in the treatment of sleep, obesity, and GH-related pathologies, such as GH deficiency.

Acknowledgments

We thank Dr Christopher J. Lelliott (Wellcome Trust Sanger Institute) for his comments and criticisms.

This work was supported by funding from the European Community's Seventh Framework Programme (FP7/2007-2013) under grant agreement no. 281854-the ObERStress European Research Council project (to M.L.), and 245009-the Neurofast project (to R.N., C.D., and M.L.), Xunta de Galicia (09CSA011208PR and XUGA CN2012/142 [to C.A.V.]); 10PXIB208164PR and 2012-CP070 (to M.L.); EM 2012/039 and 2012-CP069 (to M.L.), Junta de Andalucía (P08-CVI-03788 [to M.T.S.]), Instituto de Salud Carlos III (PI12/01814 [to M.L.]), MINECO cofunded by the FEDER Program of EU (BFU2010-16652 [to C.A.V.]); BFU2011-25021 (to M.T.S.); RN:RyC-2008-02219 and BFU2012-35255 (to R.N.); BFU2011-29102 (to C.D.). CIBER de Fisiopatología de la Obesidad y Nutrición is an initiative of Instituto de Salud Carlos III. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Disclosure Summary: Authors declare no conflict of interest.

Abbreviations

• ADX

  adrenalectomized

• ARC

  arcuate nucleus of the hypothalamus

• CORT

  corticosterone

• CREB

  cAMP response element-binding protein

• GNX

  gonadectomized

• HPX

  hypophysectomized

• ICV

  intracerebroventricular

• LHA

  lateral hypothalamus

• NPY

  neuropeptide Y

• ORX

  orchidectomized

• OVX

  ovariectomized

• OX-A

  orexin A

• pCREB

  phosphorlated CREB.