d-Lys-GHRP-6 does not modify the endocrine response to acylated ghrelin or hexarelin in humans

1. Introduction 

Ghrelin is a 28 aminoacid peptide predominantly produced by the stomach but widely expressed also in many other central and peripheral tissues (van der Lely et al., 2004, Korbonits et al., 2004, Kojima and Kangawa, 2005). In its acylated form ghrelin is a natural ligand of the previously orphan GH Secretagogue (GHS) receptor (GHS-R) type 1a (GHS-R1a) that had been shown to mediate the GH releasing action of synthetic peptidyl and non-peptidyl GHS in vitro and in vivo, in animals as well as in humans (Smith, 2005). Widespread GHS-R 1a distribution in both central and peripheral tissues has been demonstrated and the existence of GHS-R subtypes has also been hypothesized (van der Lely et al., 2004, Korbonits et al., 2004, Smith, 2005).

Besides the potent GH-releasing activity, synthetic GHS and ghrelin turned out also to exert other neuroendocrine activities at the hypothalamus/pituitary level such as stimulation of PRL and ACTH secretion (van der Lely et al., 2004, Korbonits et al., 2004, Smith, 2005). Recent reports also revealed that ghrelin/GHS also influence appetite and energy expenditure, peripheral metabolic functions, and even non-endocrine targets such as the cardiovascular system (van der Lely et al., 2004, Korbonits et al., 2004, Smith, 2005).

The physiological role of ghrelin in the control of neuroendocrine functions is still unclear and the availability of a specific antagonist would obviously represent a useful tool to define at best ghrelin function in vivo.

Further interest in selecting ghrelin antagonists also derives from the potential clinical perspective of developing new compounds able to negatively modulate appetite and food intake and therefore useful in the pharmacological approach to obesity (Horvath et al., 2003, Halem et al., 2004). BIM-28163, a full competitive antagonist of the GHS-R1a, has been shown to decrease spontaneous GH secretion, mainly lowering pulse amplitude, as well as the GH response to ghrelin but does not modulate food intake (Halem et al., 2004, Zizzari et al., 2005, Halem et al., 2005).

Since many years, d-Lys-GHRP-6 has been reported as a synthetic antagonist of the GHS-R1a, being able to antagonize the effects of various peptidyl and non-peptidyl GHS in various experimental models in vitro and in vivo (Cheng et al., 1989, Pierno et al., 2003, Duxbury et al., 2003, Glavaski-Joksimovic et al., 2003, Beck et al., 2004, Conconi et al., 2004, Fukushima et al., 2005, Kim et al., 2005, Kitazawa et al., 2005, Chen et al., 2005, Dong et al., 2006, Ohinata et al., 2006, Sibilia et al., 2006, Soares et al., 2006).

The effects of d-Lys-GHRP-6 in humans has, however, never been tested so far.

Based on the foregoing, we aimed to verify the antagonistic action of d-Lys-GHRP-6 on the neuroendocrine responses to acylated ghrelin (AG) and hexarelin (HEX), a peptidyl GHS, in humans. To this goal, the effects of d-Lys-GHRP-6 on both spontaneous and AG- or HEX-stimulated GH, PRL, ACTH and cortisol levels were studied in normal young male volunteers.

2. Subjects and methods 

Six healthy young male volunteers (age [mean±SEM]: 25.4±1.2yr; BMI: 22.3±1.0kg/m2) were studied. All the subjects gave their written informed consent to participate in the study which had been approved by an independent Ethical Committee.

All the subjects underwent the following tests (session A) in random order and at least 3 days apart:

(a)AG (1.0μg/kg iv as bolus at 0min);

(b)HEX (1.0μg/kg iv as as bolus at 0min);

(c)d-Lys-GHRP-6 (2.0μg /kg iv as bolus at 0min);

(d)d-Lys-GHRP-6 (2.0μg/kg/h iv as infusion over 120 min from −30 to +90min);

(e)AG (1.0μg/kg iv as bolus at 0min)+d-Lys-GHRP-6 (2.0μg/kg iv as bolus at 0min);

(f)AG (1.0μg/kg iv as bolus at 0min)+d-Lys-GHRP-6 (2.0μg/kg/h iv as infusion over 120min from −30 to +90min);

(g)HEX (1.0μg/kg iv as bolus at 0min)+d-Lys-GHRP-6 (2.0μg/kg iv as bolus at 0min);

(h)HEX (1.0μg/kg iv as bolus at 0min)+d-Lys-GHRP-6 (2.0μg/kg/h iv as infusion over 120min from −30 to +90min);

(i)isotonic saline (infusion from −15 to +120min+injection of 2ml as bolus at 0min);

2.1. Session B 

After the evaluation of the results of the previous session, four subject also underwent the following tests:

(a)AG (0.25μg/kg iv as bolus at 0min);

(b)AG (0.25μg/kg iv as bolus at 0min)+d-Lys-GHRP-6 (2.0μg/kg iv as bolus at 0min followed by 4.0μg/kg/h iv as infusion over 120min from 0 to +120min).

All the test have been performed in the morning starting at 08.30–09.00 after overnight fasting, 30min after an indwelling catheter had been placed into an forearm vein kept patent by slow infusion of isotonic saline.

In session A blood samples were taken every 15min from −30 up to +90min in order to assay GH, PRL, ACTH and cortisol levels.

In study B blood samples were taken every 15min from −15 up to +120min in order to assay GH levels.

Serum GH levels (ng/ml) were measured in duplicate by immunoradiometric assay (hGH IRMA CT, RADIM SpA, Pomezia, Roma). The sensitivity of the assay was 0.15ng/ml. The inter- and intra-assay coefficients of variation were 2.9–4.5% and 2.4–4.0%, respectively.

Serum PRL levels (ng/ml) were measured in duplicate by immunoradiometric assay (PRL IRMA, IMMUNOTECH distr. PANTEC, Torino). The sensitivity of the assay was 0.5ng/ml. The inter- and intra-assay coefficients of variation were 8.0% and 2.8%, respectively.

Plasma ACTH levels (pg/ml) were measured by immunoradiometric assay (ACTH, Nichols Institute Diagnostic, San Juan Capistrano, CA, USA). The sensitivity of the assay was 1pg/ml. The ranges of inter- and intra-assay coefficients of variation were 2.4–8.5% and 3.9–9.9%, respectively.

Serum cortisol levels (ng/ml) were measured in duplicate by radioimmunoassay (RIA CORTISOLO, IMMUNOTECH distrib. PANTEC, Torino). The sensitivity of the assay was 10nM. The inter- and intra-assay coefficients of variation were 5.3–9.2% and 2.8–5.8%, respectively.

All samples from an individual subject were analyzed together. The hormonal responses are expressed as mean±SEM of absolute or delta areas under curves (AUC) calculated by trapezoidal integration. The statistical analysis was carried out using non-parametric ANOVA (Friedman or Kruskall–Wallis test) and then Wilcoxon matched pairs test or Mann–Whitney U test as appropriate; correlations were carried out using the Spearman correlation coefficient.

2.2. Results 

During saline administration, no significant variations in GH and PRL levels occurred (Table 1). On the other hand, ACTH and cortisol levels showed the classical progressive decrease in morning hours.

Table 1. Mean (±SEM) GH, PRL, ACTH and cortisol ΔAUC levels after isotonic saline, acylated ghrelin (AG; 1.0μg/kg iv as a bolus), hexarelin (HEX, 1.0μg/kg iv as as bolus), d-Lys-GHRP-6 (DLG[b]; 2.0μg /kg iv as bolus), d-Lys-GHRP-6 (DLG[i]; 2.0μg/kg/h iv as infusion over 120min) alone or in coadministration

		Saline	AG	DLG[b]	DLG[i]	AG+DLG[b]	AG+DLG[i]	HEX	HEX+DLG[b]	HEX+DLG[i]
	GH (ng/ml/h)	2.2±0.5	328.1±60.6	2.1±0.4	1.9±0.2	305.0±58.8	323.9±38.4	254.3±66.5	241.5±35.6	250.9±39.7
	PRL (ng/ml/h)	32.1±2.7	91.6±22.9	31.4±2.6	31.7±1.7	89.5±2.7	89.5±9.8	57.2±2.9	55.4±5.6	54.0±2.3
	ACTH (pg/ml/h)	102.5±7.5	268.4±124.8	82.1±5.3	95.7±6.1	209.2±53.7	240.1±45.3	185.0±24.7	171.1±31.1	200.6±38.7
	Cortisol (ng/ml/h)	765.8±51.3	1099.0±83.8	772.5±72.2	764.5±39.3	1092.2±33.6	1071.2±75.0	1106.2±75.4	1126.9±106.6	1017.5±110.7

As expected, AG administration markedly increased (p<0.01) circulating GH, PRL, ACTH and cortisol levels (Table 1).

Similarly, HEX administration induced a remarkable acute increase of circulating GH (p<0.05), PRL (p<0.01), ACTH (p<0.05) and cortisol (p<0.01) secretion (Table 1).

The GH, PRL, ACTH and cortisol responses to AG were higher than those to HEX, although this difference did not attain statistical significance.

Either given as a bolus or as continuous infusion, d-Lys-GHRP-6 did not significantly modify spontaneous GH, PRL, ACTH and cortisol secretion (Table 1).

d-Lys-GHRP-6 administration as a bolus or as continuous infusion did not modify the GH, PRL, ACTH and cortisol responses to AG (Table 1).

Similarly, the GH, PRL, ACTH and cortisol responses to HEX were not modified by both as a bolus and as infusion administration of d-Lys-GHRP-6 (Table 1).

Even a very low AG dose as 0.25μg/kg iv as a bolus markedly increased (p<0.05) circulating GH (917.0±239.1ng/ml/h) levels. The GH response to this AG dose was not modified by d-Lys-GHRP-6 administration as bolus followed by continuous infusion (888.3±242.3ng/ml/h).

2.2.1. Side effects 

Four out of the six subjects after 1.0μg/kg ghrelin administration and two out of the four subjects after 0.25μg/kg ghrelin administration reported to feel hungry at the end of the testing session. Transient facial flushing was recorded in three subjects after 1.0μg/kg ghrelin administration. No side effects were reported following d-Lys-GHRP-6 and/or HEX administration.

3. Discussion 

The results of this study show for the first time that d-Lys-GHRP-6 does not affect the acute neuroendocrine responses to AG as well as to HEX, a synthetic peptidyl GHS, in humans.

d-Lys-GHRP-6 is a well-known GHS-R1a antagonist, the functional properties of which have been extensively studied both in vitro and in vivo in animals (Cheng et al., 1989, Pierno et al., 2003, Duxbury et al., 2003, Glavaski-Joksimovic et al., 2003, Beck et al., 2004, Conconi et al., 2004, Unniappan and Peter, 2004, Fukushima et al., 2005, Kim et al., 2005, Kitazawa et al., 2005, Chen et al., 2005, Dong et al., 2006, Ohinata et al., 2006, Sibilia et al., 2006, Soares et al., 2006).

In fact, an effective GHS-R1a antagonist has been initially searched for in order to develop a new tool to explore ghrelin physiology in vitro and in vivo (van der Lely et al., 2004, Halem et al., 2004). Later, further interest has been induced by evidence that ghrelin and synthetic GHS also stimulate appetite and food intake and exert independent peripheral metabolic actions (van der Lely et al., 2004, Korbonits et al., 2004, Broglio et al., 2006). These actions, in fact, would suggest the perspective of ghrelin analogues with antagonist or agonistic effect as new therapeutic approaches to obesity and/or metabolic syndrome and, as opposite conditions, to wasting syndromes.

In order to test GHS-R1a activation antagonism in vivo in humans, we studied the effects of different doses of d-Lys-GHRP-6 given as intravenous bolus or as infusion on GH, PRL, ACTH and cortisol levels either under saline infusion or in coadministration with AG or HEX. Our findings show that d-Lys-GHRP-6 alone does not modify spontaneous pituitary hormonal secretion nor the neuroendocrine responses to AG as well as to HEX.

d-Lys-GHRP-6 has been reported to inhibit GHS-induced GH release from rat pituitary cell culture (Cheng et al., 1989) but, on the other hand, not to modify spontaneous as well as GHS-induced GH secretion in rats (Cheng et al., 1997, Okimura et al., 2003). Similarly d-Lys-GHRP-6 has been reported to inhibit ghrelin-induced LH release but not to modify the GH response in goldfish (Unniappan and Peter, 2004). These data seem in contrast with other studies in animals in vivo in which other GHS-R1a antagonists, such as BIM-28163, were shown to antagonize the GH releasing effect of ghrelin and of synthetic GHS (Halem et al., 2004, Zizzari et al., 2005, Halem et al., 2005) but also with all the studies showing a significant inhibitory effect of d-Lys-GHRP-6 itself on the non-endocrine ghrelin/GHS actions in vivo (Beck et al., 2004, Conconi et al., 2004, Unniappan and Peter, 2004, Fukushima et al., 2005, Kitazawa et al., 2005, Chen et al., 2005, Dong et al., 2006, Ohinata et al., 2006, Sibilia et al., 2006).

In order to explain these discrepancies, the evidence that d-Lys-GHRP-6 affinity for the GHS-R1a is significantly lower than that of ghrelin and HEX should be carefully taken into account (Traebert et al., 2002). In fact, all the available in vivo animal studies exploring the effects of d-Lys-GHRP-6 on the non-endocrine actions of ghrelin or synthetic GHS have been performed using an antagonist/agonist concentration ratio remarkably higher (from 37:1 to 5600:1) (Traebert et al., 2002, Kitazawa et al., 2005) than those we used in the present study (from 2:1 to 4:1).

In fact, as for safety reason the absolute dose of d-Lys-GHRP-6 could not be further increased, after the negative results of the studies performed with antagonist/agonist concentration ratio of 2:1, a second study session was performed increasing the ratio to 4:1 by decreasing the absolute ghrelin dose to 0.25mcg/kg. Notably, even in this condition, though still far from antagonist/agonist concentration ratio used in animal studies, d-Lys-GHRP-6 resulted unable to modify the acute neuroendocrine actions of AG.

In conclusion, at least in our experimental conditions, d-Lys-GHRP-6 does not affect spontaneous GH, PRL and ACTH secretion and does not inhibit the acute neuroendocrine response to AG as well as to synthetic GHS in humans given at close to equimolar dose ratio. Although these negative results do not per se exclude that, even at these doses, d-Lys-GHRP-6 might have some neuroendocrine effects after prolonged treatment, or that at higher doses may be able to effectively antagonize ghrelin action in humans, nevertheless these data strongly suggest that d-Lys-GHRP-6 is unlikely to be a considered as a practical tool to explore ghrelin physiology and, hypothetically, have therapeutic perspectives in humans.