Estrogens need insulin-like growth factor I cooperation to exert their neuroprotective effects in post-menopausal women

BACKGROUND

The abrupt fall in estrogens levels during the menopausal transition may connote an hormonal state predisposing to neurodegenerative disorders, e.g. Alzheimer's disease (AD). Reportedly, the neurotrophic activity of estrogen involves an interaction with IGF-I.

AIM

To evaluate the leukocyte gene expression of progesterone receptor (PR-A/B) and interleukin 6 (IL-6), two parameters under the control of estrogens and involved in the pathogenesis of AD.

SUBJECTS

The study was conducted in non-demented women divided into two groups according to their pre- or post-menopausal state; each group being further divided into two subgroups based on their circulating levels of IGF-I (normal or low). An additional sample of AD-affected women served as a comparison group.

RESULTS

Estrogens maintained their full activity only when IGF-I levels were in the range of normalcy. On the contrary, if the concentrations of one or both hormones were reduced, estrogens were not anymore capable to control the gene expression of PR-A/B or IL-6.

CONCLUSIONS

Before administering hormone-based replacement therapy, characterization of the somatotropic function should be performed in the early phase of the menopause.

Effect of a somatostatin infusion on circulating levels of adipokines in obese women

OBJECTIVE

Changes in circulating levels of many adipocyte-derived peptides, including adipokines such as adiponectin, leptin and tumor necrosis factor alpha (TNF-α), have been reported in obesity (OB). Somatostatin (SRIF) inhibits circulating levels of adiponectin and leptin in lean (LN) subjects, but the effect of a SRIF infusion on these adipokines, including TNF-α, in OB is to date unknown.

METHODS

Ten young women (5 OB and 5 LN) were studied. All subjects underwent an infusion of SRIF (9 μg/kg/h i.v., over 60 min), with blood samples drawn prior to and at different time intervals after SRIF administration. Plasma levels of adiponectin, leptin and TNF-α were measured at each interval.

RESULTS

Basal levels of leptin and TNF-α were significantly higher in OB than LN women, whereas levels of adiponectin were significantly lower in OB than LN subjects. SRIF significantly inhibited plasma concentrations of adiponectin (at 60 min) in both OB and LN women, without affecting those of leptin and TNF-α in either group. In LN subjects, the inhibitory effect of SRIF on plasma adiponectin persisted up to 150 min, whereas SRIF infusion withdrawal in OB women resulted in a prompt restoration of basal levels of the adipokine.

CONCLUSIONS

Plasma concentrations of leptin and TNF-α, which are higher in OB than LN subjects, are unaffected by a SRIF infusion, which, in contrast, inhibits circulating levels of adiponectin in both groups, with a delayed return to the baseline secretion of the adipokine in LN subjects.

Effect of somatostatin infusion on peptide YY secretion: studies in the acute and recovery phase of anorexia nervosa and in obesity

OBJECTIVE

Changes in many gastrointestinal peptides, including the anorexigenic peptide YY (PYY), which is produced by L cells, occur in both anorexia nervosa (AN) and obesity (OB). High PYY levels are present in AN, whereas in morbid OB fasting and postprandial PYY secretion is blunted. Somatostatin (somatotropin release-inhibiting factor (SRIF)) reportedly inhibits plasma PYY concentrations in animals and healthy humans, but the effect of a SRIF infusion on spontaneous PYY secretion in AN and OB is unknown.

METHODS

A total of 18 young women, seven with acute AN (A-AN), four with AN in the recovery phase (R-AN), and seven with morbid OB, were studied. All subjects underwent an infusion of SRIF (9 μg/kg i.v./h, over 60 min), with blood samples drawn before and at different time intervals after SRIF administration. Plasma PYY levels were measured at each time point.

RESULTS

SRIF significantly inhibited plasma PYY concentrations in R-AN and OB, without affecting PYY titers in A-AN. In OB, the inhibitory effect of SRIF also persisted at 90 min. Withdrawal of SRIF infusion in R-AN resulted in a prompt restoration of basal plasma PYY levels, whereas termination of SRIF infusion in OB was followed by a slower increase of PYY titers toward baseline levels. After infusion, PYY Δ area under the curve (ΔAUC) in R-AN was significantly higher than those in A-AN and OB patients. A significant difference in PYY ΔAUC between A-AN and OB was present.

CONCLUSIONS

These results suggest the existence of a hypo- and hyper-sensitivity of L cells to the inhibitory effect of SRIF in A-AN and OB respectively.

The cholestyramine-induced decrease of PYY postprandial response is negatively correlated with fat mass in obese women

Obese patients have decreased fasting and postprandial levels of peptide YY (PYY), an anorexigenic peptide produced by the L cells of the gastrointestinal mucosa. Fatty nutrients are the most powerful stimulus for PYY release. Cholestyramine, an anion exchanger which adsorbs bile salts, reduces digestion of lipids. The aim of the present study was to investigate the effects of cholestyramine or placebo on PYY secretion in obese women administered a high-fat meal [n=8; age: 30.9±2.7 years; BMI: 47.3±3.3 kg/m2]. Postprandial PYY levels in obese women given placebo significantly increased in plasma at 30, 60, 90, and 120 min after meal ingestion. Cholestyramine administration significantly reduced postprandial PYY response at 15, 30, and 60 min. Percent fat mass (FM%) was negatively correlated with the percent increment of plasma PYY concentrations induced by meal administration at 30 min; conversely, there was a positive correlation between FM% and the percent decrement of plasma PYY concentrations induced by cholestyramine at the same time interval. These correlations failed to reach statistical significance when related to BMI. This study implies that in the obese state the altered PYY response to food consumption is a consequence of a dysfunction of L cells, which become less sensitive to the positive feedback effect of lipids.

Muscle expressions of MGF, IGF-IEa, and myostatin in intact and hypophysectomized rats: effects of rhGH and testosterone alone or combined

Myostatin and mechano-growth factor (MGF), an isoform of insulin-like growth factor-I (IGF-I), are two important regulators of muscle hypertrophy. The aim of the present study was to investigate the effects of recombinant human growth hormone (rhGH) and/or testosterone on muscle MGF/IGF-IEa/myostatin expression in intact and hypophysectomized rats treated for 15 d with 1) saline or rhGH, 2) sesame oil or testosterone, 3) saline+sesame oil, or rhGH+testosterone (first experiment) or for 7 d with saline or rhGH (second experiment). Animals were killed by decapitation 24 h or 4 d after the last injection (first or second experiment, respectively). Muscle expressions of MGF, IGF-IEa, and myostatin were determined by RT-PCR. A significant increase in the weight of gastrocnemius muscle was observed only in hypophysectomized rats treated with rhGH alone or in combination with testosterone. Administration of rhGH to hypophysectomized rats caused a marked increase in both MGF and IGF-IEa muscle mRNA levels (without any change in the muscle expression of myostatin), an effect that was abolished when testosterone was combined with rhGH. Conversely, in intact rats rhGH increased myostatin muscle mRNA levels without affecting those of MGF and IGF-IEa. Testosterone, alone or combined with rhGH, induced an inhibition of myostatin expression in the muscle of intact rats, but did not change muscle paradigms of hypophysectomized rats. In conclusion, rhGH and/or testosterone anabolic effects in the muscle are mediated by a different expression of MGF/IGF-IEa/myostatin, which is related to the pituitary function.

Menopausal transition: A possible risk factor for brain pathologic events

BACKGROUND AND OBJECTIVE

Incidence and prevalence of Alzheimer's disease (AD) are higher in postmenopausal women than in age-matched men. Since at menopause the endocrine system and other biological paradigms undergo substantial changes, we thought to be of interest studying whether (and how) the balance between some biological parameters allegedly neuroprotective (e.g. related to estrogen, dehydroepiandrosterone and CD36 functions) and others considered pro-neurotoxic (e.g. related to glucocorticoid and interleukin-6 activities) vary during lifespan in either sex in either normalcy or neurodegenerative disorders.

SUBJECTS AND METHODS

Along with this aim, we evaluated the gene expression levels of estrogen receptors (ERs), glucocorticoid receptors (HGRs), interleukin-6 (IL-6) and CD36, a scavenger receptor of class B allegedly playing a key role in the proinflammatory events associated with AD, in a population of 209 healthy subjects (73M, 106F, 20-91-year old) and 85 AD patients (36M, 49F, 65-89-year old). Results obtained were related to plasma titers of estrogens, cortisol and dehydroepiandrosterone sulfate (DHEAS). Studies were performed in peripheral leukocytes, since these cells (1) are easily obtainable by a simple blood sampling, (2) express many molecules and multiple receptors which are under the same regulatory mechanisms as those operative in the brain and (3) some of them, e.g. monocytes, share many functions with microglial cells.

RESULTS

In healthy men all the study parameters were quite stable during lifespan. In women, instead, at menopausal transition, some changes that may predispose to neurodegeneration occurred. In particular, there was (1) an up-regulation of ERs, and a concomitant increase of IL-6 gene expression, events likely due to the loss of the inhibitory control exerted by estradiol (E(2)); (2) an increase of HGR alpha:HGR beta ratio, indicative of an augmented cortisol activity on HGR alpha not sufficiently counteracted by the inhibitory HGR beta function; (3) a reduced CD36 expression, directly related to the increased cortisol activity; and (4) an augmented plasma cortisol:DHEAS ratio, widely recognized as an unfavorable prognostic index for the risk of neurodegeneration. In AD patients of both sexes, the expression of the study parameters was similar to that found in sex- and age-matched healthy subjects, thus indicating their unrelatedness to the disease, and rather a better correlation with biological events.

CONCLUSIONS

Menopausal transition is a critical phase of women's life where the occurrence of an unfavorable biological milieu would predispose to an increased risk of neurodegeneration. Collectively, the higher prevalence of AD in the female population would depend, at least in part, on the presence of favoring biological risk factors, whose contribution to the development of the disease occurs only in the presence of possible age-dependent triggers, such as beta-amyloid deposition.

The leukocyte expression of CD36 is low in patients with Alzheimer's disease and mild cognitive impairment

CD36, a scavenger receptor of class B (SR-B), helps mediate microglial and macrophage response to beta-amyloid fibrils (betaA), and seems to play a key role in the proinflammatory events associated with Alzheimer disease (AD) in many tissues. Peripheral leukocytes express many molecules and multiple receptors which undergo the same regulatory mechanisms as those operative in the brain. Thus, these cells, easily obtainable through peripheral blood sampling, may be used as a tool to investigate changes occurring in inaccessible brain areas. Based on these premises, we investigated the leukocyte expression of CD36 in 70 AD patients and in 30 subjects with mild cognitive impairment (MCI). Results were compared to those of 20 young and 40 age-matched control subjects. Leukocyte expression of CD36 was significantly reduced versus controls in both AD and MCI patients, while in young and old controls there were no age-related changes. Although preliminary, these data indicate that the reduction of CD36 expression in leukocytes is a disease-related phenomenon, occurring since the early stages of AD (MCI). Irrespective of the mechanism(s) underlying such changes, assessment of leukocyte CD36 expression might represent an useful tool to support the diagnosis of AD and to screen MCI patients candidates to develop the disease.

Early tolerance to the hypophagic effect of the cannabinoid receptor antagonist SR141716 does not impede blockade of an orexigenic stimulus

The cannabinoid CB1 receptor antagonist SR141716 (Rimonabant) is known to reduce food intake by central and peripheral mechanisms. Recently, SR141716 has been reported to block the orexigenic effect of ghrelin, a potent orexigenic peptide produced by the stomach. This study investigated whether in rats, made tolerant to the hypophagic effect of SR141716, the drug was still capable to block the orexigenic activity of another non-natural (hypothalamic) peptide, i.e., the growth hormone releasing peptide (GHRP) hexarelin, a ghrelin mimetic. In the acute experiments, each dose of SR141716 (1, 5 and 10 mg/kg i.p.) reduced food intake with respect to vehicle-treated rats, whereas hexarelin (160 microg/kg s.c.) markedly stimulated feeding. All doses of SR141716 were capable to reduce the orexigenic effect of the GHRP. A 15-day administration of SR141716 (10 mg/kg i.p.) reduced both food intake and body weight. Tolerance to the hypophagic effect of SR141716 developed within 5 days, but in contrast, body weight remained markedly below that of vehicle-treated group throughout the entire treatment period. Interestingly, despite development of tolerance to its hypophagic effect, SR141716 was capable to suppress the orexigenic effect of repeated hexarelin challenge tests performed throughout the chronic experiments. In conclusion, the results of the present study confirm and broaden the existence of a functional relationship between ghrelin and endocannabinoids in the control of food intake, and bespeak the ability of a CB1 receptor antagonist to suppress orexia caused by stimuli alien to direct stimulation of the cannabinoid system.

Orexigenic Effects of a Growth Hormone Secretagogue and Nitric Oxide in Aged Rats and Dogs: Correlation With the Hypothalamic Expression of Some Neuropeptidergic/Receptorial Effectors Mediating Food Intake

Hypothalamic neurochemical alterations in mammals underlie disturbances of food intake. There is scarce information on these topics in elderly persons; therefore, the aims of the present study were: (i) to evaluate the orexigenic effects of a growth hormone secretagogue, administered to young and old rats and dogs, alone or in combination with molsidomine, a donor of nitric oxide and (ii) to evaluate by reverse transcription-polymerase chain reaction in the whole hypothalamus of young and old rats messenger RNA levels of a wide number of anabolic and catabolic peptides, receptors, and enzymes involved in the control of feeding behavior, relating the detected titers, whenever possible, to the feeding responses to growth hormone secretagogue. In all, the results obtained strengthen the proposition that, in the hypothalamus of old rats, anti-anorexigenic compensatory mechanisms are operative, aimed at maintaining a "normal" feeding pattern. Thus, the occurrence of a primary, age-related alteration in the feeding mechanisms is unlikely.

Testosterone inhibition of growth hormone release stimulated by a growth hormone secretagogue: studies in the rat and dog

Anabolic steroids are frequently taken by athletes and bodybuilders together with recombinant human GH (rhGH), though there is some scientific evidence that the use of anabolic steroids reverses the rhGH-induced effects. Recently, we have shown that treatment with rhGH (0.2 IU/kg s.c., daily x 12 days) in the dog markedly reduced the canine GH (cGH) responses stimulated by EP51216, a GH secretagogue (GHS), evaluated after 3 and 5 daily rhGH injections, and that the inhibition was still present a few days after rhGH discontinuation. The aim of the present study was to evaluate in the dog the GH response to EP51216 (125 mug/kg i.v.) in a condition of enhanced androgenic function (i.e. acute injection or 15-day treatment with testosterone at the dose of 2 mg/kg i.m. on alternate days), and in the hypophysectomized rat the hypothalamic and hippocampal expression of ghrelin, the receptor of GHSs (GHS-R), GH-releasing hormone (GHRH) and somatostatin (SS) after specific hormonal replacement therapies (testosterone, 1 mg/kg/day s.c.; hydrocortisone, 500 mug/kg/day s.c.; rhGH, 400 mug/kg/day s.c.; 0.9% saline 0.1 ml/kg/day s.c.; x11 days). In the dog experiments, under baseline conditions, a single injection of EP51216 elicited an abrupt rise of plasma cGH. Twenty-four hours from the acute bolus injection of testosterone, C(max) and AUC(0-90) of the GHS-stimulated cGH response were significantly lower than baseline cGH response; 5 days later, there was still a significant decrease of either parameter versus the original values. Short-term treatment with testosterone markedly reduced the GHS-stimulated cGH responses evaluated during (5th bolus) and at the end (8th bolus) of testosterone treatment. Four and 8 days after testosterone withdrawal, the EP51216-stimulated cGH response was still significantly reduced when compared with that under baseline conditions. Plasma concentrations of insulin-like growth factor 1 (IGF-1) were stable until the 5th bolus of testosterone and decreased progressively in the remaining time of the testosterone treatment; 4 and 8 days from treatment withdrawal, IGF-1 levels were still suppressed. In rat studies, hypothalamic mRNA levels of GHS-R were significantly reduced by treatments with testosterone and hydrocortisone, whereas hippocampal expressions of ghrelin, GHRH and SS were reduced by rhGH replacement therapy. In conclusion, these studies show that a single administration of testosterone can abrogate the cGH response ensuing acute stimulation by a GHS; the inhibitory effect of testosterone on the cGH response to GHS is present during and even 8 days after termination of a short-lived treatment with testosterone; these events occur via a

Abuse of recombinant human growth hormone: studies in two different dog models

The search for inappropriately high growth hormone (GH) titers in plasma has been widely used to detect GH abuse, despite many shortcomings especially related to the pulsatile nature of GH secretion. Hence, the need for new anti-doping strategies. In the present study dogs were used to evaluate the ability of recombinant human GH (rhGH) to affect canine GH (cGH) release ensuing after somatostatin (SS) infusion withdrawal (SSIW) - a purported stimulus for the release of endogenous GH-releasing hormone (GHRH) - or the cGH response to administration of a GH-releasing peptide (GHRP). In the SSIW experiments, 8 beagle dogs of either gender (4-6 years old) were given a subcutaneous bolus injection of physiological saline (0.1 ml/kg) or, alternatively, rhGH (0.2 IU/kg s.c.) 60 min before the starting a continuous infusion of SS (4 microg/kg g h i.v.) of 1.5 h duration. In the dogs given a saline bolus, SSIW was followed by a 'rebound' rise in plasma cGH levels. In contrast, in dogs which had received the bolus injection of rhGH, the cGH rise elicited by SSIW was completely abrogated. In the set of experiments with a GHRP challenge, 13 dogs of either gender (3-12 years old) received the following treatment schedule at least 15 days apart: (1) a single bolus injection of rhGH (0.2 IU/kg s.c.); (2) rhGH (0.05 IU/kg s.c.) daily for 12 days; (3) rhGH (0.2 IU/kg s.c.) on alternate days for 12 days, and (4) rhGH (0.2 IU/kg s.c.) daily for 12 days. For each treatment schedule, before treatment, during treatment (24 h from the previous rhGH injection) and 1, 5 and 10 days after treatment, all dogs received an intravenous injection of a GHRP, EP51216 (125 microg/kg). In all treatments under baseline conditions, a single injection of EP51216 elicited an abrupt rise in plasma cGH. Twenty-four hours after the injection of an acute bolus of rhGH, the C(max) and AUC(0-90) of the GHRP-stimulated cGH response were significantly lower than the baseline cGH response. Five days later, there was a trend in the C(max) and AUC(0-90) towards complete restoration of the original values. One, 5 and 10 days after the end of the daily treatment with rhGH (0.05 IU/kg s.c.), no significant changes in the GHRP-stimulated cGH responses vs. the baseline GH response were recorded. In contrast, treatment with rhGH at a dose of 0.2 IU/kg s.c., on either alternate or daily administration, markedly reduced the GHRP-stimulated cGH responses evaluated after 3 and 5 rhGH injections. One day after the last rhGH injection, the EP51216-stimulated cGH response was still significantly reduced when compared with that present under baseline conditions. Five and 10 days following termination of rhGH treatment on alternate days, no significant differences in the C(max) and AUC(0-90) of the cGH responses to EP51216 were present. Differently, following the end of daily rhGH treatment, a marked inhibition in the C(max) of the cGH response to EP51216 was still present at 1 and 5 days, though not at 10 days. In conclusion, these studies show that a single administration of rhGH can abrogate the cGH response ensuing SSIW or acute stimulation by a GHRP. The inhibitory effect of rhGH on the cGH response to GHRP is present even 5 days after termination of a short-lived treatment with rhGH at a dose (0.2 IU/kg) which, in the dog, is undoubtedly lower than that used in humans for doping purposes. Extrapolation of these preclinical results to humans may pave the way for the development of a new rhGH anti-doping test.

Molsidomine antagonizes L-NAME-induced acquisition deficits in a recognition memory task in the rat

The present study was designed to investigate the role of nitric oxide (NO) on the acquisition of a recognition memory task in the rat. For this purpose, the effects on memory exerted by pre-training administration of the NO synthase inhibitor L-NAME (N(omega)-nitro-L-arginine methyl ester) and the NO donor molsidomine (N-[ethoxycarbonyl]-3-[4-morpholinosydnomine]) were assessed by using the object recognition task, a working memory paradigm based on the differential exploration of a new and familiar object. In a first dose-response study, it was found that L-NAME (10, 30, and 60 mg kg(-1), i.p.) at 30 but not at 10 mg kg(-1) disrupted animals performance, whereas the dose of 60 mg kg(-1) induced side effects. Molsidomine (2 and 4 mg kg(-1), i.p.) at 4 but not at 2 mg kg(-1), antagonized the L-NAME-induced performance deficits. These results indicate that NO is involved in the acquisition of a recognition memory task.

Plasma ghrelin concentrations in elderly subjects: comparison with anorexic and obese patients

Ghrelin, a novel endogenous ligand for the GH secretagogue receptor, has been reported to stimulate GH secretion and food intake in both humans and other animals. Interestingly, recent data indicate that ghrelin is up- and down-regulated in anorexia nervosa (AN) and obesity, which are also known to be accompanied by increased and reduced GH levels respectively. Ageing is associated with a gradual but progressive reduction in GH secretion, and by alterations in appetite and food intake. The role of ghrelin in the decline of somatotroph function and the anorexia of ageing is unknown. To investigate the influence of age on circulating levels of ghrelin, a total of 19 young and old normal weight subjects (Y-NW, n=12; O-NW, n=7), six patients with active AN (A-AN), and seven patients with morbid obesity (OB) were studied. In addition to fasting plasma ghrelin concentrations, baseline serum TSH, IGF-I and insulin levels were measured. Mean plasma ghrelin concentrations in A-AN or OB were higher and lower respectively than those present in Y-NW. Interestingly, mean plasma ghrelin concentrations in O-NW were significantly lower than those present in Y-NW and superimposable on those of OB. The mean fasting plasma ghrelin concentrations in all groups of subjects were negatively correlated with body mass index and serum insulin levels, but not with TSH and IGF-I levels. This study provides evidence of an age-related decline of plasma ghrelin concentrations, which might explain, at least partially, the somatotroph dysregulation and the anorexia of the elderly subject.

GH and cortisol rebound rise during and following a somatostatin infusion: studies in dogs with the use of a GH-releasing peptide

GH-releasing peptides (GHRPs), a class of small synthetic peptide and non-peptide compounds, act on specific receptors at both the pituitary and the hypothalamic level to stimulate GH release in both humans and other animals. GHRPs, like corticotropin-releasing hormone (CRH), also possess acute ACTH- and cortisol-releasing activity, although the mechanisms underlying the stimulatory effect of GHRPs on the hypothalamo-pituitary-adrenal (HPA) axis are still unclear. In recent years, studies in humans and other animals have provided evidence that the rebound GH rise which follows withdrawal of an infusion of somatostatin (SS) (SSIW) is due, at least in part, to the functional activation of GH-releasing hormone (GHRH) neurons of the recipient organism. Unexpectedly, in humans, SS infusion, at a dose inhibiting basal GH secretion, has been associated with an activation of the HPA axis, leading to the hypothesis that this response was mediated, at least in part, by a central nervous system ACTH-releasing mechanism activated by the SS-induced decrease in GH secretion. Interestingly, the rebound GH rise which follows SSIW was magnified by the administration, before SS withdrawal, of a GHRP, implying that the SSIW approach could also be exploited to investigate in vivo the functional interaction in the process of GH and/or ACTH/cortisol secretion between endogenous GHRH (and/or other ACTH-releasing mechanisms) and GHRPs. In the present study, six young beagle dogs were given, on different occasions, at the beginning and at the end of a 3-h i.v. infusion of SS or saline (SAL), a bolus of physiological SAL or a GHRP compound, EP51216. SSIW induced a GH rebound rise without affecting plasma cortisol concentrations, while the withdrawal of SAL infusion was ineffective on either hormone paradigm. Administration of EP51216 at the beginning of SAL infusion evoked release of both GH and cortisol, whereas EP51216 administration at the withdrawal of SAL infusion evoked somatotroph and cortisol responses which were reduced in amplitude and duration. SS infusion significantly reduced the secretion of GH elicited by EP51216 but did not affect the rise of plasma cortisol levels. Interestingly, SSIW resulted in a marked enhancement of the somatotroph and cortisol responses evoked by EP51216. The marked rise of plasma GH levels induced by the GHRP after SSIW recalled that occurring after acute combined administration of recombinant human GHRH and EP51216, implying that exogenously delivered GHRP had synergized with the endogenous GHRH release triggered by SSIW. In contrast, acute combined administration of GHRH and the GHRP induced a cortisol response not different from that induced by GHRP alone, indicating that endogenous GHRH release was not involved in the enhanced cortisol response following EP51216 administration after SSIW. Similarly, the direct involvement of endogenous CRH could be ruled out, since i.v. administration of ovine CRH after SSIW evoked cortisol peak levels not different from those evoked by CRH at the withdrawal of SAL infusion. In conclusion, enhancement of the GH response to EP51216 alone by SSIW, to an extent reminiscent of that following combined administration of GHRH and EP61216, reinforces the view that SSIW elicits release of endogenous GHRH. Further studies are indeed necessary for a better understanding of the mechanisms underlying the enhanced cortisol response, since from now on the involvement of endogenous GHRH or CRH can be ruled out.