| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Endocrinology and Metabolism, Department of Internal Medicine, University of Turin, I-10126 Turin, Italy
Address all correspondence and requests for reprints to: Ezio Ghigo, M.D., Division of Endocrinology and Metabolism, Department of Internal Medicine, University of Turin, Corso Dogliotti 14, I-10126 Torino, Italy. E-mail: ezio.ghigo{at}unito.it.
| Abstract |
|---|
|
|
|---|
Objective: The objective of the study was to evaluate the effects of AG infusion on spontaneous and stimulated gonadotropin secretion.
Design, Participants, and Intervention: In seven young healthy male volunteers (age mean ± SEM 26.4 ± 2.6 yr), we evaluated LH and FSH levels every 15 min during: 1) iv isotonic saline infusion; 2) iv saline followed by AG; LH and FSH response to GnRH (100 µg iv as a bolus), 3) alone and 4) during AG infusion; LH and FSH response to naloxone (0.1 mg/kg iv as a slow bolus), 5) alone and 6) during AG infusion.
Results: Significant LH but not FSH pulses were recorded in all subjects under saline infusion. AG infusion inhibited LH levels [area under the curve(240–480): 415.8 ± 69.7 mIU/ml·min during AG vs. 744.6 ± 120.0 mIU/ml·min during saline, P < 0.02] and abolished LH pulsatility. No change in FSH secretion was recorded. The LH and FSH responses to GnRH during saline were not affected by AG administration. However, AG inhibited the LH response to naloxone [area under the curve (120–210): 229.9 ± 39.3 mIU/ml·min during AG vs. 401.1 ± 44.6 mIU/ml·min during saline, P < 0.01]. FSH levels were not modified by naloxone alone or in combination with AG.
Conclusions: AG inhibits both spontaneous LH pulsatility and the LH response to naloxone. Because AG does not affect the LH response to GnRH, these findings indicate that the ghrelin system mediates central inhibition of the gonadal axis.
| Introduction |
|---|
|
|
|---|
Apart from its GH-releasing activity, ghrelin has received increasing attention due to its involvement in the regulation of food intake, energy expenditure, and peripheral metabolism (4). In addition, human and animal studies have demonstrated that AG is orexigenic and it increases adiposity (4, 5, 7, 8, 9, 10). There is now clear evidence that ghrelin secretion is mostly under metabolic regulation; circulating ghrelin levels are reduced in obesity and elevated during energy restriction, malnutrition, and anorexia nervosa (11).
The strict functional association between the gonadal axis and the nutritional status is widely accepted. Energy restriction is generally associated with decreased reproductive axis activity (12, 13, 14). For instance, anorexia nervosa is paradigmatically characterized by hypogonadotropic amenorrhea and hyperghrelinemia (15, 16). Indeed, recent evidence points toward a significant role of ghrelin in the negative modulation of the gonadal axis (17, 18, 19, 20, 21, 22, 23, 24). Furthermore, ghrelin suppressed LH secretion in vivo and decreased LH responsiveness to GnRH in vitro (17, 21, 24). Whereas these findings suggested that ghrelin exerts an inhibitory action at the pituitary level, other authors have reported decreased LH pulse frequency but not amplitude during infusion in adult ovariectomized rhesus monkeys. These observations suggest that ghrelin could inhibit GnRH pulse activity (25). It has also to be considered that the impact of the ghrelin system on the gonadal axis would not be restricted to the hypothalamus-pituitary level. For instance, ghrelin and its cognate receptor transcripts have been detected in rat and human gonads. In addition, ghrelin inhibited human chorionic gonadotropin-stimulated testicular testosterone secretion in rats (18). Moreover, repeated ghrelin administration induced partial delay in the timing of puberty in male rats (26). Finally, ghrelin expression has been demonstrated in human and rodent placenta, and at this level ghrelin inhibits early embryo development in vitro and pregnancy outcome in vivo (19, 20, 26).
Based on the foregoing, we sought to evaluate whether AG exerts inhibitory influence on the gonadal axis in young male volunteers and, if so, whether this action is mediated at the pituitary or the hypothalamic level. Whereas our study was ongoing, Kluge et al. (27) reported that nighttime spontaneous LH pulsatility in healthy males is delayed and inhibited after consecutive iv AG boluses. Our present findings further indicate that AG inhibits both spontaneous LH pulsatility and responsiveness to naloxone but not that to GnRH. Thus, this study is first to provide evidence that the inhibitory effects of the ghrelin system on the gonadal axis is mediated by the central nervous system (CNS).
| Subjects and Methods |
|---|
|
|
|---|
All the subjects were free of any medications for the last 3 months before the study. They had no history of depression or eating disorders or excess physical activity.
The study protocol had been approved by an independent, local ethical committee and written, informed consent was obtained from all the subjects.
Each subject underwent the following tests: 1) iv isotonic saline infusion (1000 ml of 0.9% NaCl solution) from 0 to +480 min; 2) iv isotonic saline infusion (500 ml of 0.9% NaCl solution) from 0 to +240 min followed by iv AG (1.0 µg/kg in 2 ml isotonic saline) as a bolus at +240 min and AG infusion (2.0 µg/kg·h in 500 ml isotonic saline) from +240 to +480 min; 3) GnRH (100 µg iv as a bolus at +120 min) administration during saline infusion (500 ml of 0.9% NaCl solution) from 0 to +210 min; 4) GnRH (100 µg iv as a bolus at +120 min) during AG administration (1.0 µg/kg in 2 ml isotonic saline) as a bolus at 0 min and AG infusion (2.0 µg/kg·h in 500 ml isotonic saline) from 0 to +210 min; 5) naloxone (0.1 mg/kg at +120 min) administration during saline infusion (500 ml of 0.9% NaCl solution) from 0 to +210 min; and 6) naloxone (0.1 mg/kg at +120 min) during AG administration (1.0 µg/kg in 2 ml isotonic saline) as a bolus at 0 min and AG infusion (2.0 µg/kg·h in 500 ml isotonic saline) from 0 to +210 min. A scheme of the six test sessions is represented in Fig. 1
.
|
AG was purchased from Neosystem Laboratoire (Strasbourg, France). GnRH (Relefact LH-RH) was purchased from Aventis Pharma Deutschland GmbH (Frankfurt, Germany). Naloxone (Narcan) was purchased from Sirton MediCare s.p.a. (Como, Italy).
Blood samples were taken every 15 min for the entire duration of the tests. LH and FSH levels were measured at each time point. Total testosterone was measured at baseline and at the end of the tests. In addition, total ghrelin levels were measured every 30 min during test sessions 1 and 2.
Serum LH concentrations (milliinternational units per milliliter) were measured in duplicate by immunoradiometric assay (LH immunoradiometric assay coated tube; Radim SpA, Rome, Italy). The sensitivity of the assay was 0.20 mIU/ml. The range of inter- and intraassay coefficients of variation were 7.6–13.8 and 4.6–8.6%, respectively.
Serum FSH levels (milliinternational units per milliliter) were measured in duplicate by immunoradiometric assay (FSH immunoradiometric assay CT; Radim). The sensitivity of the assay was 0.18 mIU/ml. The range of inter- and intraassay coefficients of variations were 4.0–8.8 and 3.6–7.5%, respectively.
Total testosterone levels (nanograms per milliliter; 1 ng/ml·3.467 = 1 nmol/liter) were evaluated with RIA (ICN Pharmaceuticals Inc., MP Biomedicals, Costa Mesa, CA); the sensitivity of the assay was 0.05 ng/ml; inter- and intraassay coefficients of variations were 6.8 and 15.1%, respectively.
Plasma total ghrelin levels (picograms per milliliter) were measured after extraction with reverse-phase C18 columns, by radioimmunometric assay (Phoenix Pharmaceuticals, Inc., Belmont, CA) using 125I-labeled bioactive ghrelin as a tracer and a rabbit polyclonal antibody vs. octanoylated and des-octanoylated h-ghrelin. Sensitivity was 15 pg/tube. Based on our data, the intraassay coefficient of variation range was 0.3–10.7%.
All samples from an individual subject were measured in the same assay.
Serum hormone concentrations are expressed as the mean ± SEM of absolute values or absolute areas under the curve (AUCs) calculated by trapezoidal integration.
Statistical analysis was carried out using Wilcoxon test. Levels of statistical significance were set at P < 0.05.
Computations were performed using the statistical software package SPSS (Windows version 14; SPSS, Chicago, IL).
Multiparameter deconvolution analysis was used to evaluate pulsatile FSH and LH secretion and estimate its half-life (28). Basal secretion represents the calculated time-invariant interpulse component of the release profile. The mean secretory pulse mass (MSPM) represents the mean mass of hormone released per pulse. The mean secretory pulse height (MSPH) represents the mean hormonal peak secreted per pulse. The pulsatile production rate is the product of the number of secretory bursts and the mean mass of hormone released per pulse.
| Results |
|---|
|
|
|---|
All the subjects showed normal basal LH (mean ± SEM: 3.1 ± 0.5 mIU/ml), FSH (3.5 ± 0.3 mIU/ml), and total testosterone (5.8 ± 0.7 ng/ml) concentrations.
Circulating total ghrelin levels under AG infusion reached stable supraphysiological levels until the end of the infusion (Fig. 2
).
|
AG infusion abolished LH pulsatility in five of seven subjects, whereas reduction of MSPM, MSPH, basal secretion, and half-life was recorded in the remaining two subjects for which LH pulsatility remained detectable during AG infusion (Table 1
). Therefore, AG infusion inhibited LH levels (AUC(240–480): 415.8 ± 69.7 during AG vs. 744.6 ± 120.0 mIU/ml·min during saline, P < 0.02; mean LH concentration: 1.82 ± 0.31 vs. 3.14 ± 0.52 mIU/ml, P < 0.02) (Figs. 3
and 4
).
|
|
|
At the end of test session 1, total testosterone levels were marginally decreased with respect to baseline concentrations (5.4 ± 0.6 vs. 5.8 ± 0.7 ng/ml), whereas they were significantly reduced at the end of test session 2 with respect to baseline concentrations in the same session (4.6 ± 0.6 vs. 6.7 ± 0.6 ng/ml, P < 0.05).
Gonadotropin response to GnRH during saline or AG infusion (test sessions 3 and 4)
Under saline infusion, GnRH induced a clear response of both LH (peak vs. baseline: 22.1 ± 5.4 vs. 2.8 ± 0.6 mIU/ml, P < 0.05; AUC(120–210): 1590.0 ± 374.6 mIU/ml·min) and FSH (peak vs. baseline: 8.0 ± 1.2 vs. 4.1 ± 0.6 mIU/ml, P < 0.05; AUC(120–210): 623.6 ± 84.3 mIU/ml·min). The LH and FSH responses to GnRH were not modified by the exposure to AG infusion (peaks: 22.6 ± 3.8 and 8.5 ± 1.2 mIU/ml, respectively; AUC (120–210): 1699.9 ± 281.3 and 670.4 ± 89.6 mIU/ml·min, respectively) (Fig. 5
).
|
Under saline infusion, naloxone induced a clear increase in LH secretion (peak at 135 min vs. baseline: 5.3 ± 0.8 vs. 3.2 ± 0.4 mIU/ml, P < 0.05). FSH secretion was marginally increased after naloxone administration. The exposure to AG infusion inhibited and delayed the LH response to naloxone (peak at 180 min: 3.0 ± 0.5 mIU/ml, P < 0.05; AUC(120–210): 229.9 ± 39.3 during AG vs. 401.1 ± 44.6 mIU/ml·min during saline, P < 0.01). Also, during the AG infusion, FSH secretion was not influenced by naloxone (Fig. 6
).
|
No side effect was recorded after administration of AG. Similarly, no side effect was associated with the administration of GnRH and naloxone. No medication was required and no test had to be interrupted.
| Discussion |
|---|
|
|
|---|
Several studies suggested that the ghrelin system negatively influences the gonadal axis (17, 18, 19, 20, 21, 22, 23, 24). For instance, it was previously reported that ghrelin suppresses LH pulsatility in rodent, ovine, and primate models (17, 19, 20, 21, 22, 23, 24). After repetitive ghrelin administrations, delayed onset of puberty was observed in male rats (26). In addition, it was recently described that nighttime AG bolus injections can delay and suppress LH pulse amplitude in a group of healthy volunteers (27).
It was shown that ghrelin decreases LH responsiveness to GnRH from the pituitary in vitro (23). However, in another study, ghrelin infusion decreased LH pulse frequency but not pulse amplitude in adult ovariectomized rhesus primates, suggesting that ghrelin could inhibit the GnRH pulse activity (25).
Based on this background, we aimed to verify the inhibitory action of AG on both spontaneous and stimulated gonadotropin secretion in humans and clarify whether the inhibitory influence of ghrelin on the human gonadal axis takes place at the pituitary or at the CNS level.
Our findings demonstrate that a prolonged infusion of AG quantitatively and qualitatively inhibited LH but not FSH secretion in healthy young males; in fact, AG infusion was associated with a clear inhibition of LH AUC, mean concentration, and pulsatility (in terms of pulse height and mass). Thus, these findings agree with data from Kluge et al. (27). Previous negative data showing that acute boluses of AG, even at high doses, did not modify gonadotropin secretion in either rodents or humans (1, 29, 30) are likely to reflect the very limited time of observation and the acute administration.
In contrast with in vitro data showing that ghrelin reduced the LH response to GnRH (23), our findings show that the gonadotropin responsiveness, namely LH, to their specific neurohormone is not modified by the exposure to continuous infusion of AG. These data are against the hypothesis that the inhibitory role of ghrelin takes place at the pituitary level, at least in humans.
In agreement with these assumptions, we herein show that the exposure to AG clearly inhibits the stimulatory effect of naloxone on LH secretion. It is widely accepted that naloxone represents a CNS-mediated stimulus of gonadotropin secretion (31). In fact, there is clear evidence demonstrating that opioids play an inhibitory role on the hypothalamus-pituitary-gonadal axis (32, 33, 34, 35, 36), and naloxone, an opioid antagonist, is known to trigger LH secretion by antagonizing this inhibitory opioid influence (34).
Evidence that ghrelin, at least in its acylated form, inhibits the gonadotropin response to naloxone indicates that it plays a CNS-mediated inhibitory action on the human gonadal axis. This assumption is in full agreement with previous studies in animals. Vulliémoz et al. (25) demonstrated that in the nonhuman primate the peripheral administration of ghrelin modulates the GnRH pulse generator by decreasing LH pulse frequency. Again, Furuta et al. (17) showed that intracerebroventricular ghrelin administration in ovariectomized rats inhibited LH pulsatility.
Given the well-known relevance of the central opioid system in the control of neuroendocrine functions, namely the gonadal axis, there is also evidence that specific ghrelin receptors are present at the hypothalamic and suprahypothalamic level (37, 38, 39). Thus, interplays between the ghrelin and the opioid system in the neural control of gonadotropin secretion could be considered.
It has to be emphasized that ghrelin and its receptor is expressed within the CNS and is acting as an orexigenic signal at the central level through neuropeptides such as neuropeptide Y, agouti-related protein, and orexin (20) that, in turn, are also known to play an inhibitory role in the central control of the gonadal axis (22, 40, 41). Thus, the inhibitory influence of AG on gonadotropin secretion could be mediated by these peptides.
On the other hand, Vulliémoz et al. (42) recently indicated that ghrelin inhibitory action on LH secretion may also be mediated by CRH release.
Moreover, the GnRH pulse generator governing pulsatile LH secretion is located in the mediobasal hypothalamus including the arcuate nucleus and the ventromedial hypothalamic nucleus in rats (43), in which the ghrelin receptor, GHS-R1a, is also expressed (44). Thus, direct ghrelin action on the GnRH-secreting neurons could be considered.
Interestingly, our study did not demonstrate a quantitative or qualitative effect of AG infusion on both basal and stimulated FSH secretion. Moreover, FSH response to GnRH or naloxone was not modified by AG infusion. We can speculate that LH secretion is more sensitive to AG inhibitory effect and that FSH may be sensitive to ghrelin modulation at different doses or different administration patterns.
A dissociation in LH and FSH secretion and response to GnRH during ghrelin administration has been reported in animal in vivo and in vitro studies (21, 23). In fact, ghrelin inhibits LH secretion in prepubertal male rats, gonadectomized males, and females but does not modify FSH secretion (21). Several mechanisms, such as LH and FSH different half-life, a different modulation by steroids and inhibins, and a different sensitivity to GnRH, have been proposed to explain FSH and LH different behavior in these studies.
Whatever the mechanisms underlying the influence of ghrelin on the gonadal axis may be, the inhibitory effect of a gastroenteropancreatic hormone, like ghrelin, fits well with clinical data in pathophysiological conditions. Anorexia nervosa, malnutrition, and cachexia are generally associated with hypogonadism that reflects a functional impairment of neuroendocrine mechanisms (16, 45). Metabolic factors have a major impact on ghrelin secretion regulation, and the pathophysiological conditions mentioned above are not, by chance, associated with ghrelin hypersecretion (11, 15, 46). Thus, although animal models of ghrelin and GHS-R knockout are devoid of major reproductive defects (47), it seems reasonable to hypothesize that ghrelin hypersecretion could have a role in the functional hypogonadism connoting anorexia, malnutrition, and cachexia. Interestingly, a partial resistance of ghrelin receptors could be present in anorexia nervosa because a specific reduction in GH response to ghrelin administration has been demonstrated by our group despite ghrelin hypersecretion (15). We can speculate that the restoration of ghrelin secretion and/or sensitivity could become a goal for the pharmacologic treatment of this disorder. In keeping with this assumption, it may be hypothesized that novel substances active on ghrelin receptors (48) could play a role in the recovery of gonadal axis function in anorectic patients. However, our study was performed in male volunteers, whereas anorexia nervosa is more common in female subjects. Thus, more studies are needed to focus on the long-term effects of ghrelin on reproductive function in both sexes as well as on the neuropeptides involved in the process.
In conclusion, this study demonstrates that AG inhibits both spontaneous LH pulsatility and the LH response to naloxone. Because AG does not affect the LH response to GnRH, these findings collectively indicate that the ghrelin system plays a centrally mediated inhibitory role on the gonadal axis in humans.
| Acknowledgments |
|---|
| Footnotes |
|---|
Disclosure Statement: The authors have nothing to disclose.
First Published Online June 17, 2008
1 F.L. and L.B. contributed equally as first coauthors. ![]()
Abbreviations: AG, Acylated ghrelin; AUC, area under the curve; CNS, central nervous system; GHS-R, GH secretagogue receptor; MSPH, mean secretory pulse height; MSPM, mean secretory pulse mass.
Received January 8, 2008.
Accepted June 5, 2008.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. I. Messini, K. Dafopoulos, N. Chalvatzas, P. Georgoulias, and I. E. Messinis Effect of ghrelin on gonadotrophin secretion in women during the menstrual cycle Hum. Reprod., April 1, 2009; 24(4): 976 - 981. [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 |