| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Medicine III (C.M., P.N., A.L.), Clinical Division of Endocrinology and Metabolism; Department of Clinical Pharmacology (G.S., M.W.); and Ludwig Boltzmann Institute for Experimental Endocrinology (B.B., G.G., A.L.), Medical University of Vienna, 1090 Vienna, Austria
Address all correspondence and requests for reprints to: Christina Maier, M.D., Department of Medicine III, Clinical Division of Endocrinology and Metabolism, Medical University of Vienna, Waehringer Guertel 18-20, 1090 Vienna, Austria. E-mail: christina.maier{at}akh-wien.ac.at.
| Abstract |
|---|
|
|
|---|
Atropine alone significantly reduced fasting ghrelin levels by 25%, whereas under pyridostigmine alone ghrelin levels were unaltered. Ghrelin in combination with atropine induced significantly reduced GH concentrations compared with ghrelin administration alone for both ghrelin doses, whereas ghrelin-induced GH peak concentrations and areas under the curve were not enhanced by pyridostigmine treatment. These results suggest that, in humans, fasting ghrelin concentrations might be under cholinergic control and that the cholinergic system appears to modulate ghrelin-induced GH release.
| Introduction |
|---|
|
|
|---|
The cholinergic system takes also part in the preabsorptive or cephalic phase of insulin release, which is initiated by meal ingestion and abolished by ATR (11).
Somatostatin (SRIF) acts as a functional ghrelin antagonist: SRIF infusion suppresses ghrelin concentrations in humans (12), and the GH-releasing effect of ghrelin is blunted by SRIF infusion in young healthy volunteers (13). It is also thought that SRIF mediates the modulation of GH release by cholinergic pathways (14). In healthy volunteers, muscarinic cholinergic agonists such as pyridostigmine (PD) or pilocarpine potentiate GH response to the GHRH (15, 16) and exercise (17). Furthermore, GH release to various stimuli is suppressed by the specific M1-receptor antagonist pirenzepine (18) as well as by the unspecific muscarinic antagonist ATR, the latter probably acting at a more central level (19, 20).
In a recent study in humans (21), neither PD nor the specific M1-receptor antagonist pirenzepine was able to significantly alter the GH-releasing effect of a high dose (1 µg/kg) of ghrelin. The aim of our study was to further investigate the influence of the cholinergic system on ghrelin-induced GH release, including possible regulation mechanisms at a central level. For this purpose, ATR, a cholinergic antagonist that in contrast to pirenzepine crosses the blood-brain barrier (22), and PD were administered to healthy volunteers either alone or in combination with ghrelin at two different doses. To evaluate possible cholinergic influence on ghrelin secretion, the effect of ATR and PD administration on ghrelin plasma concentrations was determined.
| Subjects and Methods |
|---|
|
|
|---|
Twelve young, healthy male volunteers, aged 27.6 ± 1.1 (mean ± SEM) yr, with a body mass index (BMI) of 24.0 ± 0.5, were studied. All of them were nonsmokers; none of them were taking any medication. The study protocol had been approved by the Ethics Committee of the University of Vienna, and all subjects had given informed consent before study entry.
Study design and schedule
The study was designed as a randomized, double-blind, placebo-controlled, crossover study. For all subjects, six study days (A-F) with at least 3-d washout intervals (mean intervals, 5.9 ± 0.3 d) were scheduled in randomized order: day A, PD (Mestinon; Hoffmann-La Roche, Basel, Switzerland) 120 mg by mouth at time point 60 and placebo (isotonic saline) iv at time point 0; day B, ATR (atropinum sulfuricum, brand name Nycomed; Nycomed, Linz, Austria) 1 mg iv at time point 0 and placebo iv at time point 0; day C, PD at 60 and ghrelin (purchased as sterile, lyophilized powder from Calbiochem-Novabiochem, Läufelingen, CH; purity 99.8%, molecular weight: 3370.9) 0.25 µg/ kg body weight (BW) at 0; day D, ATR at 0 and ghrelin 0.25 µg/ kg BW at 0; day E, PD at 60 and ghrelin 1 µg/ kg BW at 0; and day F, ATR at 0 and ghrelin 1 µg/ kg BW at 0.
Six of the 12 subjects were also tested with the two doses (0.25 and 1 µg/kg, respectively) of ghrelin alone; these additional study days were also randomized.
Ghrelin was dissolved in isotonic saline immediately before injection; volunteers as well as researchers were blinded to ghrelin/placebo dosage.
Tests began in the morning (0830-1100) after a 12-h fasting period; subjects arrived at the same time point for each study day. A plastic cannula was inserted into an antecubital vein, and blood samples were drawn at time points 60 (study days A, C, E) and 0, 15, 30, 45, and 60 (all study days) for measurements of plasma GH, ghrelin, insulin, and blood glucose. Blood pressure (measured on the upper arm) and pulse rate (finger pulse oximetry) were monitored throughout the study period with automated devices and recorded at 60 (study days A, C, E) and 0, 15, 30, 45, and 60 (all study days). An overview of the study schedule is given in Table 1
.
|
Samples for plasma hormone measurements were centrifuged immediately at 4 C and the supernatants stored at 30 C until analysis.
GH concentrations (in nanograms per milliliter) were measured in duplicate with an immunoradiometric assay (hGh IRMA; DiaSorin, Saluggia, Italy), sensitivity 0.1 ng/ml, both inter- and intra-assay coefficients of variation 46%. Cross-reactivity of the assay with the ghrelin used in the study was tested and found to be zero.
Insulin levels (µU/ml) were assayed by a commercially available RIA (Pharmacia-Upjohn, Uppsala, Sweden), and both inter- and intraassay coefficients of variation were 5%.
Plasma ghrelin (femtomoles per milliliter) was measured with a commercial RIA (Peninsula Labs, San Carlos, CA) that uses I-125-labeled bioactive ghrelin as a tracer and polyclonal antibody raised in rabbits against the C-terminal end of human ghrelin. Blood glucose was determined according to standard laboratory procedures.
Statistical analysis
For GH, areas under the curve (AUCs) were calculated from time point 0 to 60 by trapezoidal integration using Origin 5.0 (Microcal, Northampton, MA). GH AUCs and peak levels were compared by one-way ANOVA, followed by multiple t tests with Bonferroni correction using SPSS 11.5 (SPSS Inc., Chicago, IL). Blood pressure, pulse, and ghrelin, insulin, and glucose baseline vs. 15' levels and baseline vs. 30' levels were compared with ANOVA for repeated measures followed by paired Students t test for study days A-F and Wilcoxons signed rank test for study days G and H. P < 0.05 was considered significant. Results are presented as means ± SEM.
| Results |
|---|
|
|
|---|
Ghrelin concentrations at baseline were not different between the study days. ATR administration led to a decrease of ghrelin plasma concentrations that was already statistically significant at time point 15' (baseline, 186.7 ± 17.9 fmol/ml; 15', 159.6 ± 13.6 fmol/ml; P = 0.007) and further decreased to a nadir of 140.9 ± 15.7 fmol/ml at 60'. PD alone led to a small, insignificant rise of ghrelin plasma levels (baseline, 199.4 ± 30.4 fmol/ml; 15', 195.6 ± 14 fmol/ml; P = 0.863) up to 207.3 ± 19 fmol/ml at time point 60'. Ghrelin levels were significantly different between ATR and PD pretreatment at time point 15' (P = 0.033). An overview of ghrelin concentrations with ATR or PD treatment is given in Fig. 1
.
|
Ghrelin administration at 0.25 µg/ kg led to an increase in circulating ghrelin concentrations with peak values of 634 ± 22.6 fmol/ml. The higher ghrelin dosage (1 µg/ kg) induced peak values of 1959.4 ± 163.5 fmol/ml. ATR or PD treatment did not influence ghrelin plasma concentrations at both dosages.
Effect of ATR and PD on basal and ghrelin-stimulated GH concentrations
GH levels at baseline were comparable for all study days. PD administration alone led to a significant GH elevation compared with ATR administration in both peak values (PD, 6.7 ± 1.2 ng/ml; ATR, 0.15 ± 0.08 ng/ml; Pdiff = 0.002) and AUCs (PD, 194 ± 45.2 ng·min/ml; ATR, 4.6 ± 3.2 ng·min/ml; Pdiff = 0.005; Fig. 2A
).
|
Consistent results were obtained when AUCs were compared: AUCs with ghrelin at 0.25 µg/kg BW were 1565.2 ± 409.8 ng·min/ml (ghrelin alone), 1841.4 ± 311.7 ng·min/ml (PD), and 474.9 ± 211 ng·min/ml (ATR) (ATR vs. ghrelin alone, P = 0.02; PD vs. ghrelin alone, P = 1) and AUCs with ghrelin at 1 µg/kg BW were 2906 ± 165.2 ng·min/ml (ghrelin alone), 2568.3 ± 232 ng·min/ml (PD), and 1524.2 ± 311.8 ng·min/ml (ATR), respectively (ATR vs. ghrelin alone, P = 0.011; PD vs. ghrelin alone, P = 0.999). Thus, ATR pretreatment led to significantly lower GH concentrations compared with ghrelin alone. In contrast, PD did not affect GH concentrations.
An overview of GH concentrations at all study days is given in Fig. 2
.
Effect of ghrelin on insulin and glucose concentrations
Ghrelin administration at both dosages also led to a significant decrease in insulin concentrations at time point 15' vs. baseline accompanied by an increase in blood glucose concentrations, which was not significant with 1 µg/kg BW ghrelin. ATR or PD treatment alone did not significantly change insulin or glucose concentrations. Also, ghrelin-induced changes in insulin levels were not influenced by ATR or PD; the effect on glucose concentrations, however, was more heterogenous. An overview of the glucose and insulin results is given in Table 2
.
|
PD administration significantly reduced (baseline 75.5 ± 3.2 min1 vs. 15' 60.6 ± 2.2 min1; P < 0.001) and ATR significantly enhanced (baseline 77.2 ± 3.9 min1 vs. 15' 87.7 ± 4.2 min1; P = 0.025) pulse rate. ATR did not lead to a significant alteration of systolic (126.7 ± 2.9 mm Hg baseline vs. 125.5 ± 3.4 mm Hg 15'; P = 0.717) or diastolic blood pressure (63.6 ± 3.9 mm Hg baseline vs. 68 ± 3.3 mm Hg 15'; P = 0.18), whereas PD significantly reduced systolic (127.2 ± 3.5 mm Hg baseline vs. 118.5 ± 3.3 mm Hg 15'; P = 0.001) but not diastolic (64 ± 2.4 mm Hg baseline vs. 60.5 ± 3.1 mm Hg 15'; P = 0.149) blood pressure. ATR and PD effects were not significantly influenced by ghrelin treatment.
| Discussion |
|---|
|
|
|---|
It has been shown that ghrelin levels are decreased by SRIF infusion in humans (12) and that SRIF is produced in the stomach mucosa. Therefore, it would be possible that the effect of ATR on ghrelin release was mediated by gastric SRIF. However, this explanation seems rather unlikely, because it has been shown in rodents that gastric SRIF release is unaffected by vagotomy (24). The observation that PD (which is assumed to act by lowering somatostatinergic tone) was unable to change ghrelin levels suggests that the physiological somatostatinergic tone might not be involved in the regulation of ghrelin concentrations during fasting.
During the preparation for this manuscript, a paper describing the effect of the muscarinic antagonist pirenzepine and PD on ghrelin secretion has been published (25). In this study, pirenzepine reduced ghrelin levels to the same amount as described here. In contrast to our data, PD was able to significantly increase ghrelin concentrations with a maximum at 120 min after PD administration. A possible explanation for this discrepancy could be that in our study ghrelin concentrations, after the 12-h fasting period, might already have reached maximal values at baseline. However, because absolute ghrelin concentrations at baseline were not reported in the paper, a direct comparison is not possible.
In addition to reducing ghrelin plasma concentrations, ATR also significantly reduced ghrelin-induced GH release in this study. In a recent study on young, healthy volunteers (21), no significant effect of manipulation of the cholinergic system with either PD or pirenzepine on the ability of ghrelin to induce GH release was found. Thus, the finding of a significant effect of cholinergic blockade with ATR on ghrelin-induced GH release is in contrast to these data. The fact that we used ATR instead of pirenzepine might account for this discrepancy. ATR, in contrast to pirenzepine, is able to cross the blood-brain barrier. Recently, it has been shown that in vivo the effect of ghrelin on GH release is mediated via the hypothalamic releasing hormone GHRH rather than directly via the pituitary (26, 27). Thus, one possible explanation for the discrepancy between the effectiveness of ATR and the ineffectiveness of pirenzepine would be that the interplay between the cholinergic system and ghrelin occurs at a central level that is not reached by pirenzepine.
This would not explain however why PD was unable to enhance ghrelin-induced GH release, even at a low dose of ghrelin in combination with a PD dose leading to a significant elevation of GH levels when administered alone (Fig. 2A
). There are several possible explanations for the striking discrepancy between the effectiveness of cholinergic blockade and the inability of cholinergic enhancement to modify ghrelin-induced GH release. According to general belief, the main influence of PD on GH release is via lowering somatostatinergic tone in the median eminence of the hypothalamus (14). It has been shown that SRIF infusion blunts ghrelin-induced GH release (13). Thus, one would expect that upon lowering SRIF influence, ghrelin effect on GH release would be enhanced. Therefore, the results of this and the above-mentioned studies suggest that normal SRIF tone might be too low to control ghrelin actions on GHRH/GH release. Alternatively, the nonadditive effect of ghrelin and PD could be attributed to the fact that both substances act on the same pathway. Indeed, the common view that PD acts mainly via lowering SRIF tone has been challenged by the use of a GHRH antagonist that almost abolished PD-induced GH release in healthy humans (28).
We also measured ghrelin plasma levels after administration of different doses of ghrelin. The peak plasma ghrelin levels (634.6 ± 22.6 fmol/ml) achieved with the lower ghrelin dosage (0.25 µg/ml) were higher than the upper range (447.8 fmol/ml) of fasting ghrelin levels reported in a recent population study in adults using the same RIA for plasma ghrelin determination (29). Thus, it can be assumed that even those plasma levels are supraphysiological for healthy humans and maybe lower doses of ghrelin than the 1 µg/kg used in most studies up to now would be more appropriate for studies on ghrelin physiology.
Ghrelin seems to suppress insulin release from pancreatic ß-cells (30, 31), and it has been shown repeatedly that insulin levels decrease after ghrelin administration in rodents (31) and humans (21, 32). Ghrelin appears also to influence gluconeogenesis in the liver (33). Significant reduction of insulin levels accompanied by elevation of glucose levels at time point 15' was also seen in this study at a ghrelin dose of 0.25 µg/kg BW, whereas at 1 µg/kg ghrelin insulin concentrations were significantly decreased, but glucose concentrations showed only a trend to increase. The interrelation among ghrelin, insulin, and glucose appears to be complex and needs to be studied in a setting with more frequent blood sampling and more ghrelin doses. The time interval of 15 min chosen in this study does not allow drawing conclusions on the actual order of glucose rise and insulin drop.
Taken together, the data from our study provide evidence that interplay exists between ghrelin and the cholinergic system in humans. It has shown that cholinergic blockade with ATR suppresses ghrelin plasma concentrations and modulation of the cholinergic system with ATR, but not PD, influences the ability of different ghrelin dosages to induce GH release.
| Footnotes |
|---|
Abbreviations: ATR, Atropine; AUC, area under the curve; BMI, body mass index; BW, body weight; PD, pyridostigmine; SRIF, somatostatin.
Received April 6, 2004.
Accepted June 16, 2004.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. Maier, M. Riedl, G. Vila, P. Nowotny, M. Wolzt, M. Clodi, B. Ludvik, and A. Luger Cholinergic Regulation of Ghrelin and Peptide YY Release May Be Impaired in Obesity Diabetes, September 1, 2008; 57(9): 2332 - 2340. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Vila, C. Maier, M. Riedl, P. Nowotny, B. Ludvik, A. Luger, and M. Clodi Bacterial Endotoxin Induces Biphasic Changes in Plasma Ghrelin in Healthy Humans J. Clin. Endocrinol. Metab., October 1, 2007; 92(10): 3930 - 3934. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |