Gastric GDF15 levels are regulated by age, sex, and nutritional status in rodents and humans.

AIM: Growth differentiation factor 15 (GDF15) is a stress response cytokine that has been proposed as a relevant metabolic hormone. Descriptive studies have shown that plasma GDF15 levels are regulated by short term changes in nutritional status, such as fasting, or in obesity. However, few data exist regarding how GDF15 levels are regulated in peripheral tissues. The aim of the present work was to study the variations on gastric levels of GDF15 and its precursor under different physiological conditions, such as short-term changes in nutritional status or overfeeding achieved by HFD. Moreover, we also address the sex- and age-dependent alterations in GDF15 physiology. METHODS: The levels of gastric and plasma GDF15 and its precursor were measured in lean and obese mice, rats and humans by western blot, RT-PCR, ELISA, immunohistochemistry and by an in vitro organ culture system. RESULTS: Our results show a robust regulation of gastric GDF15 production by fasting in rodents. In obesity an increase in GDF15 secretion from the stomach is reflected with an increase in circulating levels of GDF15 in rats and humans. Moreover, gastric GDF15 levels increase with age in both rats and humans. Finally, gastric GDF15 levels display sexual dimorphism, which could explain the difference in circulating GFD15 levels between males and females, observed in both humans and rodents. CONCLUSIONS: Our results provide clear evidence that gastric GDF15 is a critical contributor of circulating GDF15 levels and can explain some of the metabolic effects induced by GDF15.

Uroguanylin: a new actor in the energy balance movie.

Uroguanylin (UGN) is a potential target in the fight against obesity. The mature protein is released after enzymatic cleavage from its natural precursor, proUGN. UGN is mostly produced in the gut, and its production is regulated by nutritional status. However, UGN is also produced in other tissues such as the kidneys. In the past, UGN has been widely studied as a natriuretic peptide owing to its involvement in several different pathologies such as heart failure, cancer and gastrointestinal diseases. However, recent studies have suggested that UGN also acts as a regulator of body weight homeostasis because it modulates both food intake and energy expenditure. This ultimately results in a decrease in body weight. This action is mediated by the sympathetic nervous system. Future studies should be directed at the potential effects of UGN agonists in regulating body weight in human obesity.

FNDC5 is produced in the stomach and associated to body composition.

The fibronectin type III domain-containing protein 5 (FNDC5) discovered in 2002 has recently gained attention due to its potential role in protecting against obesity. In rat, no data exist regarding FNDC5 production and regulation in the stomach. The aim of the present work was to determine the expression of FNDC5 in the rat stomach and its potential regulation by body composition. The present data shows FNDC5 gene expression in the gastric mucosa. Immunohistochemical studies found FNDC5 immunopositivity in chief cells of gastric tissue. By the use of three different antibodies FNDC5 was found expressed in gastric mucosa and secreted by the stomach. The rate of gastric FNDC5 secretion parallels the circulating levels of FNDC5. The body fat mass increase after intervention with high fat diet coincided with a decrease in the secretion of FNDC5 from the stomach and a diminution in the FNDC5 circulating levels. In summary, the present data shows, for the first time, the expression of FNDC5 in the stomach of rats and its regulation by body composition, suggesting a potential role of gastric FNDC5 in energy homeostasis.

Secreted factors derived from obese visceral adipose tissue regulate the expression of breast malignant transformation genes.

BACKGROUND/OBJECTIVES: Obese adipose tissue, especially the visceral depot, exhibits altered production of several molecules that could have a role on the initiation/promotion of breast cancer development. The aim of this work was to evaluate the effect of excess adipose tissue and its secreted factors on the expression of genes involved in the early steps of tumor promotion on the mammary gland. SUBJECTS AND METHODS: Carcinogenesis-related gene expression was evaluated in mammary gland tissue from female diet-induced obese (DIO) Sprague-Dawley rats and circulating leukocytes isolated from a group of breast cancer diagnosed and non-diagnosed obese women and compared with their normal weight counterparts. In addition, the human non-tumoral mammary epithelial cell line MCF10A was treated in vitro with the visceral (retroperitoneal adipose tissue (RPAT)) or subcutaneous adipose tissue (SAT) secretome and with rising concentrations of the lipid peroxidation by-product 4-hydroxynonenal (4-HNE). RESULTS: DIO rats were classified as susceptible to DIO (DIO-S) or partially resistant to DIO (DIO-R) according to the maximum fat mass gain of the lean group as a cut-off. As compared with lean and DIO-R, the DIO-S group showed a higher fat mass and lower lean mass. The anatomical characteristic of DIO-S was correlated with differential expression of cellular proliferation (ALDH3A1 and MYC) and antioxidant and DNA protection (GSTM2, SIRT1), and tumor suppression (TP53, PTEN, TGFB1) genes. Remarkably, this carcinogenesis-related gene expression pattern was reproduced in MCF10A treated with the RPAT secretome from DIO-S rats and with the lipid peroxidation by-product 4-HNE. Moreover, this pattern was also detected in leukocytes from obese women compared with normal weight women without evidence of breast cancer. CONCLUSIONS: Lipid peroxides secreted by the obese visceral adipose tissue could be among the relevant factors that promote changes involved in the early steps of tumor development in mammary gland. These changes can be detected even before histological alterations and in circulating leukocytes.

Uroguanylin levels in intestine and plasma are regulated by nutritional status in a leptin-dependent manner.

PURPOSE: Uroguanylin (UGN) is a 16 amino acid peptide produced mainly by intestinal epithelial cells. Nutrients intake increases circulating levels of prouroguanylin that is processed and converted to UGN to activate the guanylyl cyclase 2C receptor (GUCY2C). Given that the UGN-GUCY2C system has been proposed as a novel gut-brain endocrine axis regulating energy balance, the aim of the present study was to investigate the regulation of UGN protein levels in duodenum and circulating levels in lean and obese mice under different nutritional conditions and its potential interaction with leptin. METHODS: Swiss, C57BL/6 wild-type and ob/ob male adult mice under different nutritional conditions were used: fed ad libitum standard diet (control); 48 h fasting (fasted); 48 h fasting followed by 24 h of feeding (refed); and fed high-fat diet (45 %) during 10 weeks. In addition, peripheral leptin administration was performed. Intestinal uroguanylin expression was studied by Western blot analysis; plasma levels were measured by ELISA. RESULTS: Food deprivation significantly reduced plasma UGN levels, which were correlated with the lower protein levels of UGN in duodenum. These effects were reverted after refeeding and leptin challenge. Consistently, in ob/ob mice UGN expression was decreased, whereas leptin treatment up-regulated UGN levels in duodenum in these genetically modified mice compared to WT. Diet-induced obese mice displayed increased UGN levels in intestine and plasma in comparison with lean mice. CONCLUSIONS: Our findings suggest that UGN levels are correlated with energy balance status and that the regulation of UGN by nutritional status is leptin-dependent.

Drug development strategies for the treatment of obesity: how to ensure efficacy, safety, and sustainable weight loss.

The prevalence of obesity has increased worldwide, and approximately 25%-35% of the adult population is obese in some countries. The excess of body fat is associated with adverse health consequences. Considering the limited efficacy of diet and exercise in the current obese population and the use of bariatric surgery only for morbid obesity, it appears that drug therapy is the only available method to address the problem on a large scale. Currently, pharmacological obesity treatment options are limited. However, new antiobesity drugs acting through central nervous system pathways or the peripheral adiposity signals and gastrointestinal tract are under clinical development. One of the most promising approaches is the use of peptides that influence the peripheral satiety signals and brain-gut axis such as GLP-1 analogs. However, considering that any antiobesity drug may affect one or several of the systems that control food intake and energy expenditure, it is unlikely that a single pharmacological agent will be effective as a striking obesity treatment. Thus, future strategies to treat obesity will need to be directed at sustainable weight loss to ensure maximal safety. This strategy will probably require the coadministration of medications that act through different mechanisms.

The brain-stomach connection.

The stomach-brain connection has been revealed to be one of the most promising targets in treating obesity. The stomach plays a key role in the homeostatic mechanism implicating stomach-brain communication regulated under neural and hormonal control. The present review explores specific topics related to gut-brain interactions focus on the stomach-brain connection through the different known systems implied in energy balance control as ghrelin, and nesfatin. Moreover, novel mechanisms for energy balance regulation involving gastric-brain communication are described including the role of the gastric intracellular mTOR/S6K1 pathway mediating the interaction among ghrelin, nesfatin and endocannabinoid gastric systems to modulate metabolism.

The vagus nerve as a regulator of growth hormone secretion.

The growth hormone (GH) axis is mainly regulated by the growth hormone releasing hormone (GHRH) and somatostatin (SS) hypothalamic peptides. Nevertheless, since ghrelin peptide was discovered as the natural ligand for growth hormone secretagogue receptor (GHS-R), the mechanism of GH regulation has acquired a new dimension. It was described that ghrelin possesses a relevant effect inducing GH secretion when it is administered peripherally. A role of the vagus nerve mediating ghrelin action has been described although this effect is not understood. Under this context the main objective of this work was to determine the possible involvement of the vagus in the somatotroph axis regulation. The results in this manuscript show that animals with a disruption of the vagus connection presented lower basal IGF-1 and GH levels, a decreased GH response to peripheral GHRH administration and a marked diminution in the GH response to peripheral and central ghrelin treatments. In addition, vagotomized animals showed a down-regulation of GHRH mRNA in the arcuate nucleus of the hypothalamus and a down-regulation in both GHRH and GHS receptors' mRNA at the pituitary level. In conclusion, the present work reveals that the vagus nerve is crucial in growth hormone regulation and essential for the full GH-releasing effect of ghrelin.

Macronutrients act directly on the stomach to regulate gastric ghrelin release.

BACKGROUND: Ghrelin is a gastric secreted hormone deeply implicated in meal initiation and body weight regulation. This peptide is a peripheral orexigenic hormone with a nutritional status-dependent regulation showing a pre-pandrial rise and post-prandial fall pattern. A wide variety of studies have tested the effect of meal different nutrient composition over stomach mucosa ghrelin content and plasmatic ghrelin levels; nevertheless, few and non-conclusive data exist about the direct action of macronutrients on the stomach in order to regulate ghrelin secretion. The recent identification of taste receptors or chemoreceptors in the stomach mucosa would reinforce this paradigm. AIMS: To investigate the individual effect of different macronutrients (l-glutamine, lipids, and glucose) over gastric ghrelin secretion by using an in vitro gastric explants model. RESULTS: L-glutamine and intralipid emulsion act locally in the stomach decreasing ghrelin secretion, while no effect was found after glucose exposure. CONCLUSIONS: These results show for the first time that macronutrients, and specially amino acids and lipids, act directly in the stomach in order to regulate gastric ghrelin release. Consequently, the chemosensory capacity of the stomach, until now restricted to the oral cavity or intestine, is demonstrated.

Peripheral endocannabinoid system-mediated actions of rimonabant on growth hormone secretion are ghrelin-dependent.

The somatotroph axis is a crucial pathway regulating metabolism. Despite the fact that the endocannabinoid system has been also revealed as a potent modulator of energy homeostasis, little information is available concerning a putative interaction between these two systems. The aim of the present study was to determine the in vivo effects of the blockade of the cannabinoid receptor type 1 (CB1) over growth hormone (GH) secretion using the CB1 antagonist rimonabant. The results obtained show that the blockade of the CB1 peripheral receptor by i.p. injection of rimonabant significantly inhibited pulsatile GH secretion. Similarly, it was found that this injection significantly decreased ghrelin-induced GH secretion without any effect on growth hormone-releasing hormone (GHRH)-induced GH discharge. In situ hybridisation showed that the peripheral blockade of CB1 did not affect hypothalamic somatostatin mRNA levels; however, GHRH mRNA expression was significantly decreased. The blockade of the vagus nerve signal by surgical vagotomy eliminated the inhibitory action of rimonabant on GHRH mRNA and consequently on GH. On the other hand, the central CB1 blockade by i.c.v. rimonabant treatment was unable to reproduce the effect of peripheral blockade on GHRH mRNA, nor the GH response to ghrelin. In conclusion, the data reported in the present study establish, from a physiological point of view, the existence of a novel mechanism of GH regulation implicating the action of the cannabinoid receptor on the somatotroph axis.

Age, sex, and lactating status regulate ghrelin secretion and GOAT mRNA levels from isolated rat stomach.

Ghrelin is a stomach derivate peptide involved in energy homeostasis regulation, and ghrelin O-acyltransferase (GOAT) is the enzyme responsible for ghrelin acylation. Puberty is a period characterized by profound changes in the metabolic requirements and notable variations of sexual hormone levels. On the other hand, the weaning process is a fundamental modification of the diet, which implicates several adaptations of the gastrointestinal tract physiology. Until now the direct secretion of ghrelin by the stomach in these conditions, without interferences from other organs, has never been studied. The main objective of this article was to investigate how the stomach modulates ghrelin production and secretion as well as GOAT expression on these periods of life. Gastric ghrelin secretion is regulated through postnatal life in an independent way of gastric expression and circulating levels of this hormone. The present work shows a strong regulation of gastric ghrelin secretion by estrogens. The weaning strongly regulates gastric ghrelin secretion. Animals subjected to delayed weaning present a lower body weight than the corresponding controls. For the first time, it is shown that a noticeable decrease in circulating levels of testosterone and estrogens is associated with delay of weaning. GOAT mRNA levels in the stomach are strongly regulated by age, breastfeeding, and testosterone. In conclusion, the stomach itself regulates ghrelin and GOAT production to adapt the organism to the metabolic requirements demanded through each stage of life.

Growth hormone and somatostatin directly inhibit gastric ghrelin secretion. An in vitro organ culture system.

Ghrelin is a 28-amino-acid hormone produced mainly by the stomach which strongly promotes food intake. It is the only known peripheral orexigenic hormone that induces the release of GH. Ghrelin has been proposed as a link between the enteric system and central regulation of energy balance and growth. Although it has recently been the focus of extensive study, the secretion mechanism is not yet well characterized. The aim of this study was to test the direct effect of hormones from the somatotropic axis on ghrelin release directly from the stomach. To this end, an organ culture model of gastric tissue explants from rat donors was used. These stomach explants were incubated in 6 well plates for 1, 2, and 3 h after treatment with either GH, GHRH, SS or IGF-I, all them at 10(-6) M. After incubation, the medium was collected and the amount of ghrelin secreted by the gastric tissue was measured by radioimmunoassay. It was observed that GH and SS significantly decreased gastric ghrelin secretion, while GHRH and IGF-1 had no effect on the present model. These results would confirm the capacity of GH and SS to act directly upon gastric level, inhibiting ghrelin secretion in vitro.

Perinatal overfeeding in rats results in increased levels of plasma leptin but unchanged cerebrospinal leptin in adulthood.

OBJECTIVE: To study the effect of perinatal programming and overfeeding on the hypothalamic control mechanisms of food intake in adult rats. DESIGN: Neonatal programming effects on body weight, food intake, central and peripheral leptin levels, hypothalamic neuropeptides, leptin receptors and central leptin responsiveness in adult rats. MEASUREMENTS: Plasma and cerebrospinal fluid (CSF) leptin levels were analyzed using radioimmunoassay. Neuropeptide mRNA levels were analyzed using in situ hybridization. Leptin receptor mRNA levels were analyzed using reverse transcriptase-polymerase chain reaction. RESULTS: Perinatally overfed rats growing up in small litters (SL) maintain their obese and hyperleptinemic phenotype in adulthood. However, leptin levels in CSF are abnormally low considering the plasmatic hyperleptinemia. In contrast to the already reported changes in perinatally overfed juvenile rats, perinatally overfed adult rats did not show any alteration in the expression of leptin receptor isoforms and evaluated neuropeptides. Moreover, SL adult rats showed a normal sensitivity regarding the inhibitory effect of intracerebroventricular leptin administration on food intake. CONCLUSION: Perinatal overfeeding does not induce alterations in either the anorectic response to central leptin administration or expression of leptin receptors and neuropeptides in adulthood. The leptin resistance to peripheral leptin in SL adult rats may be related to impaired leptin transport across the blood-brain barrier.

Central obestatin administration does not modify either spontaneous or ghrelin-induced food intake in rats.

The isolation of ghrelin unveiled a new system implicated in food intake regulation. The recently isolated hormone obestatin derives from the same precursor of ghrelin and seems to perform opposite actions. It could be part of a dual system connecting gut and brain to regulate energy homeostasis. The ability of intracerebroventricular administration of obestatin to modify food intake was evaluated. Obestatin had no effect on spontaneous food intake in both ad libitum and food restricted rats. The obestatin injection was not able to antagonize the ghrelin-stimulated increase in food intake either. In conclusion, the present work does not support a role for obestatin on the regulation of food intake in any model studied.

A possible role of neuropeptide Y, agouti-related protein and leptin receptor isoforms in hypothalamic programming by perinatal feeding in the rat.

AIM/HYPOTHESIS: Perinatal overfeeding predisposes humans and rats to obesity and diabetes in later life. One classical model for studying the effect of early feeding is manipulation of the size of rat litters. Rats growing up in small litters gain more weight than rats growing up in normal-sized litters. Interestingly, these obese rats maintain this phenotype in adulthood. Conversely, rats raised in large litters show a delay in growth and a decrease in body weight. The aim of this work was to assess the hypothalamic control mechanisms of food intake regulated by perinatal feeding. METHODS: Leptin levels were analysed using RIA. Leptin receptor mRNA levels were analysed using RT-PCR. Neuropeptide mRNA levels were analysed using in situ hybridisation. RESULTS: Perinatally overfed neonatal male rats exhibited hyperleptinaemia and a decrease in hypothalamic mRNA levels of the long isoform of the leptin receptor (OB-Rb), explaining their leptin resistance. Moreover, this obese model showed an increase in the mRNA expression of cocaine- and amphetamine-regulated transcript, neuropeptide Y and agouti-related protein in the hypothalamic arcuate nucleus (ARC). In contrast, perinatally underfed neonatal male rats with hypoleptinaemia showed an increase in hypothalamic mRNA of the short isoforms of the leptin receptor. Furthermore, they exhibited an increase in expression of neuropeptide Y and agouti-related protein in the ARC. CONCLUSIONS/INTERPRETATION: Rats overfed during early postnatal life show a leptin-resistant state mediated by down-regulation of the hypothalamic OB-Rb. These data, together with the increased expression of neuropeptide Y and agouti-related protein in specific neurons in the ARC, might indicate the existence of regulated programming in this nucleus and may provide a new aetiopathogenic concept in susceptibility to obesity.

[Role of ghrelin in the pathophysiology of eating behaviour]. / Papel de la ghrelina en la fisiopatología del comportamiento alimentario.

Ghrelin, the endogenous ligand for GHS-R, was isolated from rat stomach, although other tissues exist expressing ghrelin, such as pituitary, hypothalamus, placent, ovary, testes, etc. It was showed that ghrelin is implicated in GH secretion, in vivo and in vitro. There are direct evidences that proof that ghrelin administration induces GH secretion. There are in vivo data, showing ghrelin as a most potent GH secretor than GHRH. Evidences exist of ghrelin actions in the regulation of food intake and energy homeostasis. Ghrelin has a clear role in the differents pathologies such as obesity, anorexia and bulimia.

Orexin A suppresses in vivo GH secretion.

BACKGROUND/AIMS: Orexins (OXs) are a newly described family of hypothalamic neuropeptides. Based on the distribution of OX neurons and their receptors in the brain, it has been postulated that they could play a role in the regulation of neuroendocrine function. GH secretion is markedly influenced by nutritional status and body weight. To investigate the role OX-A plays in the neuroregulation of GH secretion we have studied its effect on spontaneous GH secretion as well as GH responses to GHRH and ghrelin in freely moving rats. Finally, we also assessed the effect of OX-A on in vitro GH secretion. METHODS: We administered OX-A (10 microg, i.c.v.) or vehicle (10 microl, i.c.v.) to freely moving rats. Spontaneous GH secretion was assessed over 6 h with blood samples taken every 15 min. RESULTS: Administration of OX-A led to a decrease in spontaneous GH secretion in comparison with vehicle-treated rats, as assessed by mean GH levels (means+/-s.e.m. 4.2+/-1.7 ng/ml vs 9.4+/-2.2 ng/ml; P<0.05), mean GH amplitude (3.6+/-0.5 ng/ml vs 20.8+/-5.6 ng/ml; P<0.01) and area under the curve (848+/-379 ng/ml per 4 h vs 1957+/-458 ng/ml per 4 h; P<0.05). In contrast, OX-A failed to modify in vivo GH responses to GHRH (10 microg/kg, i.v.) although it markedly blunted GH responses to ghrelin (40 microg/kg, i.v.) (mean peak GH levels: 331+/-71 ng/ml, vehicle, vs 43+/-11 ng/ml in OX-A-treated rats; P<0.01). Finally, OX-A infusion (10(-7), 10(-8) or 10(-9) M) failed to modify in vitro basal GH secretion or GH responses to GHRH, ghrelin and KCl. CONCLUSIONS: These data indicate that OX-A plays an inhibitory role in GH secretion and may act as a bridge among the regulatory signals that are involved in the control of growth, nutritional status and sleep regulation.

Cellular distribution and regulation of ghrelin messenger ribonucleic acid in the rat pituitary gland.

Ghrelin, a 28-amino-acid acylated peptide, strongly stimulates GH release and food intake. In the present study, we found that ghrelin is expressed in somatotrophs, lactotrophs, and thyrotrophs but not in corticotrophs or gonadotrophs of rat pituitary. Persistent expression of the ghrelin gene is found during postnatal development in male and female rats, although the levels significantly decrease in both cases from pituitaries of 20-d-old rats onward, but at 60 d old, the levels were higher in male than female rats. This sexually dimorphic pattern appears to be mediated by estrogens because ovariectomy, but not orchidectomy, increases pituitary ghrelin mRNA levels. Taking into account that somatotroph cell function is markedly influenced by thyroid hormones, glucocorticoids, GH, and metabolic status, we also assessed such influence. We found that ghrelin mRNA levels decrease in hypothyroid- and glucocorticoid-treated rats, increase in GH-deficient rats (dwarf rats), and remain unaffected by food deprivation. In conclusion, we have defined the specific cell types that express ghrelin in the rat anterior pituitary gland. These data provide direct morphological evidence that ghrelin may well be acting in a paracrine-like fashion in the regulation of anterior pituitary cell function. In addition, we clearly demonstrate that pituitary ghrelin mRNA levels are age and gender dependent. Finally, we show that pituitary ghrelin mRNA levels are influenced by alteration on thyroid hormone, glucocorticoids, and GH levels but not by fasting, which indicates that the regulation of ghrelin gene expression is tissue specific.

Hypothalamic levels of NPY, MCH, and prepro-orexin mRNA during pregnancy and lactation in the rat: role of prolactin.

Pregnancy and lactation provide excellent models of physiological hyperphagia and hyperprolactinemia. To identify possible factors associated with the increased feeding in these situations, we measured hypothalamic mRNA levels of three orexigenic neuropeptides--NPY, MCH, and orexins--in nonpregnant, pregnant, and lactating rats by in situ hybridization. NPY mRNA content in the arcuate nucleus was significantly increased during pregnancy and lactation. However, MCH and prepro-orexin expression was decreased in both states. 48 or 72 h of fasting in pregnant and lactating rats further elevated NPY mRNA levels and increased the low MCH mRNA content. Surprisingly, no effect was observed in prepro-orexin mRNA levels. Finally, we investigated the possible effect of high PRL levels on these orexigenic signals using a model of hyperprolactinemia induced by pituitary graft. NPY mRNA content was unchanged, but MCH and prepro-orexin mRNA levels were significantly decreased. Our results suggest that the increased NPY expression might be partly responsible for the hyperphagia observed during pregnancy and lactation. MCH and prepro-orexin may be involved in the adaptation of other homeostatic mechanisms and their decreased levels in these physiological settings could be mediated by the elevated circulating PRL levels.

Influence of thyroid status and growth hormone deficiency on ghrelin.

OBJECTIVE: To assess whether some of the alterations in energy homeostasis present in thyroid function disorders and GH deficiency could be mediated by ghrelin. DESIGN: To assess the influence of thyroid status on ghrelin, adult male Sprague-Dawley rats were treated with vehicle (euthyroid), amino-triazole (hypothyroid) or l-thyroxine (hyperthyroid). The influence of GH on ghrelin was assessed in wild-type (control) and GH-deficient (dwarf) Lewis rats. Evaluation of gastric ghrelin mRNA expression in the stomach was carried out by Northern blot. Circulating levels of ghrelin were measured by radioimmunoassay. RESULTS: Hypothyroidism resulted in an increase in gastric ghrelin mRNA levels (euthyroid: 100+/-3.2% vs hypothyroid: 127.3+/-6.5%; P<0.01), being decreased in hyperthyroid rats (70+/-5.4%; P<0.01). In keeping with these results, circulating plasma ghrelin levels were increased in hypothyroid (euthyroid: 124+/-11 pg/ml vs hypothyroid: 262+/-39 pg/ml; P<0.01) and decreased in hyperthyroid rats (75+/-6 pg/ml; P<0.01). Using an experimental model of GH deficiency, namely the dwarf rat, we found a decrease in gastric ghrelin mRNA levels (controls: 100+/-6% vs dwarf: 66+/-5.5%; P<0.01) and circulating plasma ghrelin levels (controls: 124+/-12 pg/ml vs dwarf: 81+/-7 pg/ml; P<0.01). CONCLUSION: This study provides the first evidence that ghrelin gene expression is influenced by thyroid hormones and GH status and provides further evidence that ghrelin may play an important role in the alteration of energy homeostasis and body weight present in these pathophysiological states.