| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of Endocrinology, Diabetes, and Nutrition (A.M.A., M.O.W., H.R., C.S., J.S., M.M., A.F.H.P.), and Clinical Chemistry and Pathobiochemistry (F.H.P.), Charité-University Medicine Berlin, Campus Benjamin Franklin, 12200 Berlin, Germany; Department of Clinical Nutrition (A.M.A., M.O.W., H.R., C.S., J.S., M.M., A.F.H.P.), German Institute of Human Nutrition Potsdam-Rehbruecke, 14558 Nuthetal, Germany; and Medical Department-Innenstadt (B.O.), University Hospital Munich, 80337 Munich, Germany
Address all correspondence and requests for reprints to: Mohammad Ayman Arafat, M.D., Department of Endocrinology, Diabetes, and Nutrition, Charité-University Medicine Berlin, Campus Benjamin Franklin, Hindenburgdamm 30, 12200 Berlin, Germany. E-mail: ayman.arafat{at}charite.de.
| Abstract |
|---|
|
|
|---|
Objective: The objective of this study was to evaluate the putative role of the hypothalamic pituitary axis in glucagons suppressive effect on ghrelin secretion.
Design, Subjects, and Methods: Prospectively, we studied the endocrine and metabolic responses to im glucagon administration in 22 patients (16 males; age, 2168 yr; body mass index, 28.1 ± 1.1 kg/m2) with a known hypothalamic-pituitary lesion and at least one pituitary hormone deficiency. Control experiments were performed in 27 healthy subjects (15 males; age, 1965 yr; body mass index, 25.5 ± 0.9 kg/m2).
Results: The suppression of ghrelin by glucagon measured as area under the curve240min was significantly greater in controls when compared with patients (P < 0.01). Although there was a significant decrease in ghrelin in controls (P < 0.001), ghrelin was almost unchanged in patients (P = 0.359). Changes in glucagon, glucose, and insulin levels were comparable between both groups.
Conclusions: We show that the hypothalamic-pituitary axis plays an essential role in the suppression of ghrelin induced by im glucagon administration. Glucagon significantly decreases ghrelin levels in healthy subjects. However, in the absence of an intact hypothalamic-pituitary axis, this effect was abolished. The mechanisms responsible for our observation are unlikely to include changes in glucose or insulin levels.
| Introduction |
|---|
|
|
|---|
Glucagon is a 29-amino-acid peptide hormone processed from proglucagon in pancreatic
-cells. Although the main physiological role of glucagon is to maintain glucose homeostasis as a major counterregulatory hormone of insulin, there is growing evidence that prandial secretion of glucagon might also play a role in the control of meal termination and satiation as demonstrated in rats (6, 7). The precise mechanism by which glucagon controls spontaneous feeding remains uncertain. We have shown recently that glucagon induces a remarkable decrease in ghrelin, a 28-amino-acid gastric peptide known to regulate feeding behavior and adiposity (8). This effect could not be explained by changes in glucose or insulin concentrations (8), and published data do not support a direct inhibitory effect of glucagon on ghrelin-producing cells in the stomach (9).
The aim of the present study was to evaluate the possible role of central mechanisms in the effect of glucagon on ghrelin secretion in humans. We hypothesized that the suppressive effect of glucagon on ghrelin may be modulated at the brain-hypothalamus-pituitary level. To test our hypothesis, we investigated ghrelin, glucagon, insulin, glucose, GH, and cortisol responses to im glucagon in patients with a known hypothalamic-pituitary lesion as well as in healthy subjects.
| Subjects and Methods |
|---|
|
|
|---|
Investigations were performed in 22 consecutively recruited patients [16 men and six women; age, 2168 yr; body mass index (BMI), 28.1 ± 1.1 kg/m2] with a known hypothalamic-pituitary dysfunction and at least one pituitary hormone deficiency (Table 1
). In brief, six of these patients had pituitary macroadenomas with suprasellar extension, 14 patients had traumatic brain injury, one patient had a meningioma with a hypothalamic involvement, and another one had HIV infection. In five patients, the tumor was removed via a transsphenoidal approach, one patient underwent radiotherapy, and another patient underwent a therapy with cabergoline. The major exclusion criteria included a history of diabetes mellitus, any current inflammatory or malignant disease, and pregnancy. All patients were on stable replacement therapy with thyroid hormone, hydrocortisone, or sex steroids as appropriate. None of the patients was on GH replacement therapy.
|
All subjects gave written informed consent to participate in the study, which was approved by the hospitals Ethical Committee. The subjects got a full medical history, a physical examination, and had height and weight recorded, from which BMI was derived.
Glucagon test
Subjects were asked to skip their medication at the morning of the test. The test started at 0830 h after an overnight fast and 30 min after an indwelling catheter had been placed into an antecubital vein. Glucagon was administered im (1 mg for subjects with a body weight < 90 kg and 1.5 mg for subjects with a body weight > 90 kg) at 0830 h, and the subjects remained supine until the end of the test. Serum and EDTA plasma samples were taken at 30, 0, 30, 60, 120, 180, and 240 min relative to the glucagon injection. Blood samples were kept frozen at 80 C until assayed.
Hormone assays
Capillary blood glucose was measured using the glucose oxidase method (glucometer Biosen 5130, EKF-diagnostic, Magdeburg, Germany). Insulin was measured with an ELISA [Mercodia, Uppsala, Sweden; inter- and intraassay coefficients of variation (CV) were 3.6 and 4%, respectively]. Paired measurements of fasting glucose and insulin were used to derive estimates of insulin resistance using the homeostasis model assessment (HOMA) as appropriate. Plasma glucagon levels were assessed in duplicate with a RIA using I125-labeled glucagon as a tracer and antibody raised in rabbits against glucagon (DPC Biermann, Bad Nauheim, Germany; intra- and interassay CV were 4.8 and 8.6%). Serum total ghrelin was quantified using a RIA (Phoenix Pharmaceuticals, Mountain View, CA; intra- and interassay CV were 5.3 and 13.6%) as previously described (8). Serum GH concentrations were determined by a commercially available chemiluminescent immunometric assay (Diagnostic Products Corporation, Los Angeles, CA; lower detection limit, 0.05 ng/ml; inter- and intraassay CV were 6.2 and 6.5%). Serum cortisol concentrations were determined by chemiluminescent immunometric assay (Diagnostic Products Corporation; CV were 10 and 8.8%).
Statistical analyses
Statistical analyses were performed using SPSS version 12 (SPSS, Chicago, IL). All data are expressed as mean ± SEM unless stated otherwise. Baseline characteristics were compared using two-tailed Students t test for unpaired values if the data were normally distributed. In case of skewed data, the nonparametric Kruskal-Wallis-Test was used. Shapiro-Wilk-Test was used to test for normal distribution. P < 0.05 was regarded as statistically significant. The baseline value was calculated as the mean of the 30- and 0-min values. Serial changes in glucagon, ghrelin, glucose, and insulin concentrations after glucagon administration were analyzed using ANOVA for repeated measures. Changes were compared with baseline using Students t test for paired analysis. All significances are two-sided, and P < 0.01 was regarded as statistically significant (as corrected by Bonferroni for multiple testing). The integrated areas under the curve (AUC) calculated by the trapezoid method were used to compare the time courses of patients and controls.
| Results |
|---|
|
|
|---|
|
|
|
|
In controls, changes in ghrelin concentrations at 60 min still proved to be significant after statistical correction for changes in glucose and insulin (P < 0.01). Moreover, the changes in AUC240-ghrelin still proved to be significant after statistical correction for changes in AUC240-insulin (P < 0.01).
Thirty minutes after im glucagon administration, glucose levels showed a maximal increase followed by a decrease to baseline level after 120 min in both patients [86.4 ± 2.4 (baseline), 130.6 ± 4.1 (30 min), and 85.8 ± 5.2 (120 min) mg/dl; P < 0.001] and controls [89.2 ± 2.9 (baseline), 140.2 ± 5.6 (30 min), and 89.3 ± 7.1 (120 min) mg/dl; P < 0.001] (Fig. 1
). The AUC240-glucose was comparable in patients and controls (262.6 ± 6.4 vs. 260.5 ± 6.3; P = 0.811) (Fig. 3
).
Insulin levels showed a similar increase with a peak after 30 min followed by a decrease toward baseline level after 120 min in both patients [10.3 ± 1.6 (baseline), 47 ± 4.6 (30 min), and 16.3 ± 3.5 (120 min) mU/liter; P < 0.01] and controls [8.1 ± 2.2 (baseline), 40.6 ± 6.4 (30 min), and 12.7 ± 2.9 (120 min) mU/liter; P < 0.01] (Fig. 1
). The AUC240-insulin was comparable in patients and controls (567.6 ± 54.5 vs. 673 ± 68.5; P = 0.41) (Fig. 3
).
To investigate whether changes in GH or cortisol might be responsible for the effect of glucagon on ghrelin, we measured the time courses of both hormones in controls. Glucagon elicited an increase in GH [3.2 ± 1.1 (baseline) vs. 13.9 ± 2.5 (peak, 180 min) µg/liter; P < 0.01] and cortisol concentrations [459.1 ± 30.4 (baseline) vs. 632.4 ± 41.3 (peak, 180 min) nmol/liter; P < 0.01] (Fig. 4
). However, no remarkable increase in the concentration of both hormones was seen during the first 2 h after the administration of glucagon, during which time ghrelin levels reached their nadir.
|
| Discussion |
|---|
|
|
|---|
Moreover, in disconcordance with the previously described effect of gender on ghrelin secretion (10, 11), the baseline ghrelin levels did not significantly differ in respect to gender. In our cohort, men and women had an equal BMI (28.4 ± 1.2 for male patients vs. 27.2 ± 2.6 kg/m2 for female patients, P = 0.68; and 24.6 ± 0.9 for male controls vs. 26.5 ± 1.6 kg/m2 for female controls, P = 0.48) and were of the same age (47.7 ± 3.5 yr for male patients vs. 43.2 ± 6.3 yr for female patients, P = 0.51; and 42.3 ± 3.8 yr for male controls vs. 35.6 ± 3.8 yr for female controls, P = 0.23). However, compared with the women included in the previously published studies, our women had a higher BMI and were older. This may explain why we did not find a difference in baseline ghrelin levels between males and females. Moreover, in both of our groups there were more men than women (six females vs. 16 males in the patients and 12 females vs. 15 males in the controls), which makes a direct comparison of the baseline ghrelin levels between both sexes difficult. Anyway, the ghrelin response to glucagon administration did not significantly differ in respect to gender in both study groups.
Taken together, a direct hypothalamus-mediated impact of glucagon on ghrelin secretion can be assumed. Our study demonstrates for the first time that glucagons suppressive effect on ghrelin may be exerted at brain-hypothalamus-pituitary level because an intact hypothalamic-pituitary axis was necessary for a maximal impact of glucagon on ghrelin levels.
Appetite is regulated in a highly complex manner, and various central and peripheral factors such as ghrelin are involved. It is expected that the understanding of these mechanisms may help to find an effective treatment for the control of body weight (1). The arcuate nucleus of the hypothalamus and the dorsal vagal complex seem to be the most important central nervous system regions directly regulating food intake (2), and obesity is a common long-term result of hypothalamic damage in adults with space-occupying lesions of the hypothalamic-pituitary region (12).
Previous studies have shown that both exogenous and endogenous pancreatic glucagon controls spontaneous meal size in rats (3, 7, 13, 14). In humans, a decrease in meal size after iv infusion of physiological doses of glucagon in men has also been reported (15). The sites of origin, the afferent and efferent mechanisms underlying these effects, have not been clearly identified. According to the findings of Geary et al. (16), glucagon is suggested to act in the liver by the induction of satiety signal that is transmitted to the brain by the hepatic branch of the abdominal vagus. However, this was not supported by studies demonstrating a similar inhibitory effect of hepatic portal and intracardiac glucagon infusions on feeding (17). Moreover, there was no reduction in glucagons satiating potency after antagonism of peripheral muscarinic receptors, and the effect of ip injected glucagon on satiety remained intact even after hepatic vagotomy (16, 18, 19, 20). This complements the findings of Jensen et al. (21), who observed that elevated levels of proglucagon-derived peptides are associated with an abrupt onset of profound anorexia and adipsia in rats with transplantable glucagonoma, an effect that is mediated neither by the neuropeptide Y-ergig system nor by a stimulation of the vagus. Thus, the contribution of the vagus to glucagon-induced satiety remains uncertain.
Indeed, ghrelin might be involved in mediating the glucagon effects on appetite regulation. The increase in plasma glucagon in our study (maximum, 245.3 ± 14 pg/ml; approximately 5-fold) was similar to its increase in some physiological states such as in response to hypoglycemia (1.5- to 7.3-fold) as reported by others (22, 23). Moreover, hypoglucagonemia might be partly responsible for the hyperphagia described in streptozotocine rats (24). Furthermore, hypoglucagonemia seen in type 1 diabetes may be responsible for the hyperphagia described in those patients. Finally, obese individuals known to have a further reduction in ghrelin concentrations after gastric bypass surgery (25) tended to have an increase in glucagon levels (26).
Our findings offer a possible model wherein reduced ghrelin in hyperglucagonemic states plays a key role as an efferent signal in mediating the already described glucagon impact on food intake.
The decrease in ghrelin level in our control subjects was comparable with the decrease reported by Hirsh et al. (27) after the administration of a 10-fold higher dose of glucagon. They found a 26% decrease in ghrelin after im glucagon (0.1 mg/kg) as well as after iv glucagon (0.03 mg/kg) in children. Therefore, the glucagon-induced effects on ghrelin do not seem to be dependent on the dose of glucagon or on its route of administration. However, the im or iv administration of a lower dose of glucagon (<0.01 mg/kg) might induce smaller effects.
To our knowledge, no information exists to date about the mechanisms underlying the glucagon-induced reduction of ghrelin levels. Glucagon-binding sites have been identified in multiple tissues, including brain stem and hypothalamus (28, 29, 30), and intracerebroventricular administration of glucagon has been shown to potently suppress food intake in rats (31). Moreover, besides its ability to receive signals from the periphery via the brain stem, the arcuate nucleus is located at the base of the hypothalamus containing an area known to exhibit a permeable blood-brain barrier, which facilitates exposure to circulating factors like glucagon (32).
As a limitation, the study size does not allow a further stratified analysis for patients with singularly defined hypothalamic-pituitary abnormalities. A regulatory feedback link between GH and ghrelin has been suggested (8, 33), and almost all of our patients were GH deficient (18 of 22). However, our controls showed no remarkable increase in GH concentrations during the first 2 h, when ghrelin levels reached their nadir. In addition, baseline ghrelin levels were comparable in both patients and controls (Table 2
). Furthermore, looking separately at the four patients with an intact GH axis, glucagon failed to induce a decrease in ghrelin levels [188.1 (baseline) vs. 231.6 (30 min), 198.5 (60 min), 193.5 (120 min), 195.1 (180 min), and 178.5 (240 min) pg/ml]. This is also supported by a recently published study showing an intact postprandial ghrelin regulation in patients with GH deficiency (12).
Another limitation of the study is that the exact extension of the lesions via imaging data is not available in all patients. However, it is likely that most of the patients had both hypothalamic and pituitary lesions due to the kind of damage (traumatic brain injury and macroadenoma with suprasellar extension). Comparing both groups (patients with macroadenoma and those with traumatic brain injury), no differences in the ghrelin responses to glucagon were noticed (data not shown). Anyway, further studies addressing this issue are needed.
As a possible mechanism, the previously described glucagon-induced hypothalamic somatostatin release might be involved in the centrally mediated glucagon-induced effect on ghrelin (34). This is supported by the observation that activation of somatostatin receptors remarkably inhibits ghrelin secretion as reported by some other studies (35, 36).
In conclusion, glucagon seems to act centrally to induce a reduction in ghrelin concentration in healthy humans. These findings provide a possible explanation of the glucagon impact on satiety and may be of potential diagnostic or therapeutic use for the treatment of certain states presenting with hyperghrelinemia.
| Footnotes |
|---|
Abbreviations: AUC, Area(s) under the curve; BMI, body mass index; CV, coefficient(s) of variation; HOMA, homeostasis model assessment.
Received January 31, 2006.
Accepted June 13, 2006.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |