Ghrelin potentiates growth hormone secretion driven by putative somatostatin withdrawal and resists inhibition by human corticotropin-releasing hormone

CONTEXT: Ghrelin is a 28-amino acid, Ser(3)-octanoylated peptide that stimulates GH secretion in vivo and in vitro. Beyond the capability of ghrelin to synergize with GHRH, little is known about multipeptide modulation of ghrelin's actions in humans. OBJECTIVE: The objective of this study was to test the hypothesis that ghrelin can stimulate GH secretion in the absence or presence of somatostatin withdrawal (induced by l-arginine infusion) and stress-like drive by CRH. DESIGN: This was a randomized, double-blind, placebo-controlled, cross-over interventional study. SETTING: This study was performed at an academic medical center. PARTICIPANTS: Nine healthy postmenopausal women not receiving sex hormones were studied. INTERVENTIONS: Subjects were given an iv infusion of saline and/or l-arginine or human CRH, followed by a bolus iv injection of ghrelin. OUTCOME MEASURES: The outcome measures were pulsatile GH secretion quantified by repetitive blood sampling, immunochemiluminometry, and deconvolution analysis. RESULTS: Consecutive saline/ghrelin infusion increased pulsatile GH secretion from 2.7 +/- 1.0 (saline/saline; mean +/- sem) to 20 +/- 5.0 microg/liter.3 h (P < 0.01). The magnitude of the effect of l-arginine/saline was comparable at 20 +/- 4.5 microg/liter.3 h (P < 0.01). In contrast, sequential l-arginine/ghrelin evoked true synergy of GH release (93 +/- 14 microg/liter.3 h; P = 0.003 vs. l-arginine alone and P = 0.008 vs. ghrelin alone). Human CRH did not affect GH responses to saline/saline (3.9 +/- 1.1 microg/liter.3 h), saline/ghrelin (19 +/- 3.3 microg/liter.3 h), l-arginine/saline (16 +/- 2.7 microg/liter.3 h), or l-arginine/ghrelin (90 +/- 13 microg/liter.3 h). CONCLUSIONS: Assuming that l-arginine reduces somatostatin outflow, we infer that ghrelin can activate hypothalamo-pituitary pathways that are both dependent upon and independent of somatostatinergic restraint even in the face of a strong stress-related signal.     

Influence of human corticotropin-releasing hormone and adrenocorticotropin upon spontaneous growth hormone secretion.

Hormones of the hypothalamo-pituitary-adrenocortical (HPA-) axis are considered to be of physiological and clinical relevance in regulating spontaneous growth hormone (GH) secretion. To further investigate interdependencies between both systems, we studied the effects of adrenocorticotropin [ACTH(1-24)] and human corticotropin-releasing hormone (h-CRH) upon spontaneous GH secretion in 10 male volunteers. Administration of 1 microgram ACTH (1-24), 10 micrograms h-CRH or saline (control: CTL) every hour from 9.00 to 6.00 p.m. resulted in significant differences of cortisol secretion during the entire observation period (8.00 a.m.-3.00 a.m.) between the three groups (p less than 0.001, Friedman two-way ANOVA). Mean area under the time course curve (AUC) values (+/- SEM) for cortisol expressed as ng x 1,000 x min/ml showed also significant differences between the three treatments from 8.00 a.m. to 3.00 a.m.: CTL 64.0 +/- 6.4, ACTH(1-24) 178.5 +/- 9.4 (p less than 0.01, Wilcoxon test), h-CRH 88.5 +/- 5.6 (p less than 0.01). The main portion of cortisol was released during daytime from 8.00 a.m. to 11.00 p.m., where the most significant differences in the AUC values emerged: CTL 59.6 +/- 5.8, ACTH(1-24) 171.5 +/- 8.8 (p less than 0.01, Wilcoxon test), h-CRH 80.2 +/- 5.1 (p less than 0.01). With regard to GH secretion, significant differences became obvious between the three treatments during daytime from 8.00 a.m. to 11.00 p.m. and the sleep-related period from 11.00 p.m. to 3.00 a.m. (p less than 0.01 and p less than 0.02, Friedman two-way ANOVA).(ABSTRACT TRUNCATED AT 250 WORDS)     

Central activin administration modulates corticotropin-releasing hormone and adrenocorticotropin secretion

A broad and diffuse neuronal network conveys information reflecting the state of the internal and external environment to the neurosecretory hypothalamus. Recently, we identified an inhibin-beta A- (I beta A) immunoreactive terminal field within the CRF-rich portion of the dorsomedial paraventricular nucleus which originates from a cell group in the commissural portion of the nucleus of the solitary tract (NTS). The NTS receives baroreceptor input, somatosensory input via the spinosolitary tract, and sensory information from the oral, thoracic, and abdominal cavities and, thus, is positioned to serve as a primary relay for visceral sensory inputs to neurons critical to the function of the hypothalamic-pituitary-adrenal (HPA) axis. Although these NTS cells contain multiple putative transmitters, we present evidence that activin, an inhibin-beta A dimer, plays a modulatory role in HPA axis function via facilitation of CRF release. First, intraventricular injection of activin-A (0-3 nmol), but not the related inhibin heterodimer, evoked dose-related 1.7- to 2.8-fold elevations of circulating ACTH levels in male rats. Second, analysis of hypophysial-portal plasma after bilateral paraventricular nucleus microinfusion of activin-A revealed a dose-related facilitation of CRF secretion up to 4-fold above preinjection levels which was unaccompanied by changes in arginine vasopressin levels. Finally, activin-A also enhanced CRF secretion from neonatal hypothalamic cells in primary culture with an EC50 dose of approximately 0.25 nM. Overall, these observations provide evidence of both an anatomical and a pharmacological substrate for activin-mediated central modulation of HPA axis function.     

Involvement of hypothalamic somatostatin in the suppression of growth hormone secretion by central corticotropin-releasing factor in conscious male rats

The role of central corticotropin-releasing factor (CRF) in the regulation growth hormone (GH) secretion was studied in freely moving conscious male rats with indwelling intra-atrial and intracerebroventricular (i.c.v.) cannulae. GH measurements in blood samples obtained every 20 min from 10.00 to 14.00 h in control animals injected with saline either intravenously (i.v.) or into the lateral cerebral ventricle revealed that spontaneous GH secretion was pulsatile, and occurred regularly at around 12.00 h. When ovine CRF (10 micrograms) was injected i.c.v., spontaneous GH secretion was inhibited (mean plasma GH [11.20-13.00 h]: 20 +/- 7 ng/ml vs. control: 126 +/- 22 ng/ml, p less than 0.01). In contrast, the intravenous injection of CRF (10 micrograms) did not affect spontaneous GH secretion (mean plasma GH [11.20-13.00 h]: 162 +/- 25 ng/ml vs. control: 193 +/- 31 ng/ml). This GH suppressive action of central CRF was blocked by the i.v. injection (0.5 ml) of antisomatostatin serum (AS), but not of normal sheep serum (NS), (mean plasma GH [11.20-13.00 h]: NS + CRF: 15 +/- 2 ng/ml vs. AS + CRF: 202 +/- 30 ng/ml, p less than 0.01). The mean plasma GH value [11.20-13.00 h] in animals receiving AS and CRF was not significantly different from those in animals receiving saline (i.v.) together with AS. These results suggest a potential inhibitory role of central CRF in the regulation of spontaneous GH secretion in the rat which is mediated by the stimulation of hypothalamic somatostatin.     

Prolactin stimulates rat hypothalamic corticotropin-releasing hormone and pituitary adrenocorticotropin secretion in vitro.

In rats hyperprolactinemia increases corticotropin-releasing hormone (CRH) concentration and secretion in hypophysial portal blood and the serum concentration of adrenocorticotropic hormone (ACTH). To determine whether the stimulatory effect of prolactin (PRL) on CRH and ACTH in vivo is exerted directly on the hypothalamus, hypothalamic explants and primary anterior pituitary cell cultures from adult male and female rats were used. Hypothalami explanted from male and female rats were preincubated during 90 min and treated for 30 min with rat PRL (rPRL) at concentrations of 10(-8), 10(-7), and 10(-6) M (about 200, 2,000, and 20,000 ng/ml, respectively), corticosterone at concentrations of 10(-7), 10(-6), and 10(-5) M (about 35,350 and 3,500 ng/ml, respectively), ACTH at concentrations ranging from 10(-10) to 10(-7) M (0.46, 4.6, 46, and 460 ng/ml, respectively), and graded concentrations of testosterone or estradiol. Concentrations of immunoreactive CRH (iCRH) were measured by radioimmunoassay. rPRL at 10(-6) M stimulated iCRH secretion by 360 and 400% of the basal iCRH output (about 14 pg/hypothalamus), respectively, from hypothalami explanted from male and female rats. ACTH and corticosterone did not suppress rPRL (10(-6) M) induced iCRH secretion. Corticosterone at the concentration of 10(-6) M potentiated rPRL (10(-6) M) induced iCRH secretion in hypothalami explanted from male, but not female rats. Gonadal steroids had no effect either on the basal or rPRL (10(-6) M) stimulated iCRH secretion, with the exception of estradiol which augmented the response to 10(-6) M rPRL by about fivefold, but only at the concentration of 10(-8) M (about 2.7 ng/ml).(ABSTRACT TRUNCATED AT 250 WORDS)     

Thyrotropin-releasing factor-induced adrenocorticotropin secretion is mediated by corticotropin-releasing factor

TSH-releasing factor (TRF), administered into the lateral cerebroventricle of adult male rats, elevated plasma concentrations of ACTH, epinephrine, and norepinephrine. TRF given iv was devoid of these activities. The CRF receptor antagonist, alpha-helical CRF9-41 (CRF9-41) given iv suppressed the TRF-induced increase in ACTH, but did not alter TRF-induced changes in plasma catecholamines. Intravenous administration of CRF antiserum totally blocked TRF-induced elevation of plasma ACTH concentrations. CRF receptor antagonists administered icv attenuated CRF-induced, but not TRF-induced elevation of plasma concentrations of ACTH, epinephrine, and norepinephrine. It is concluded from these results that TRF acts within the central nervous system to stimulate ACTH release through a CRF-dependent mechanism.     

Regulation of ghrelin secretion by somatostatin analogs in rats

OBJECTIVE: Ghrelin is a hormone present in the plasma in two forms: octanoylated and des-octanoylated ghrelin. In pathophysiological conditions such as Prader-Willi syndrome and ghrelinoma, elevated ghrelin plasma levels are associated with pathological obesity. Clinical studies have shown that somatostatin downregulates ghrelin plasma levels in healthy volunteers. The aim of this study was to investigate the effects of two somatostatin analogues, SOM230 and octreotide, on ghrelin secretion in rats. METHODS: Ghrelin secretion was either unstimulated or stimulated by overnight fasting. Treatment with SOM230 and octreotide was either acute (s.c. injection 1 h before blood sampling) or prolonged (continuous s.c. infusion via 14-day osmotic minipumps). RESULTS: Acute treatment with octreotide dose-dependently inhibited unstimulated and stimulated secretion of total and active ghrelin. SOM230 (30 microg/kg) inhibited active ghrelin in fasted rats. Lower doses had no effect. After 7 days of treatment, active ghrelin was strongly inhibited by both compounds in fasted animals, with a stronger effect for octreotide. Lower inhibition was achieved in fed rats. After 14 days, the inhibition with octreotide in fasted rats was lower and SOM230 had no effect. Somatostatin receptor expression analysis in the rat glandular stomach revealed a predominant sst(1) and sst(2) expression, low expression of sst(3) and sst(4), and hardly detectable sst(5) mRNA expression. CONCLUSIONS: Somatostatin analogues may be useful for the inhibition of physiologically elevated ghrelin plasma levels. This inhibition appears to be mediated by sst(2) receptors in the rat, and desensitizes after 14 days of treatment.     

Effect of metyrapone pretreatment on adrenocorticotropin secretion induced by corticotropin-releasing hormone in normal subjects and patients with Cushing's disease

To examine the influence of endogenous cortisol on the ACTH response to CRH, we compared ACTH secretion during CRH tests before and after metyrapone administration in 9 normal subjects and 12 patients with Cushing's disease. The administration of 4.5 g metyrapone (750 mg, orally, every 4 h) resulted in a decrease in basal (pre-CRH) plasma cortisol levels and an increase in basal plasma ACTH levels in both normal subjects and Cushing's patients. The pretreatment with metyrapone significantly blunted the increase in plasma cortisol levels and markedly enhanced ACTH secretion after iv injection of 100 micrograms human CRH. The peak ACTH levels during CRH test before and after metyrapone administration were 8 +/- 1 and 58 +/- 8 pmol/L, respectively, in normal subjects (P less than 0.01) and 26 +/- 5 and 50 +/- 11 pmol/L, respectively, in Cushing's patients (P less than 0.05). Although the basal and peak ACTH levels as well as delta ACTH (peak ACTH - basal ACTH) during the CRH test before metyrapone administration were significantly higher in Cushing's disease patients than in normal subjects (P less than 0.01), no such difference was observed between the 2 groups after metyrapone administration. The results clearly indicate that the endogenous cortisol levels greatly influence the ACTH response to CRH, and that the CRH test as commonly performed does not allow a correct evaluation of potential responsiveness of normal pituitaries and Cushing's adenomas to CRH.     

Interactions between tumor necrosis factor-alpha, hypothalamic corticotropin-releasing hormone, and adrenocorticotropin secretion in the rat

We studied the effects of tumor necrosis factor-alpha (TNF alpha), a macrophage-derived pleiotropic cytokine produced during the inflammatory/immune response, on the function of the hypothalamic-pituitary-adrenal (HPA) axis of the rat. Intravenous injections of TNF alpha stimulated plasma ACTH and corticosterone secretion in a dose-dependent fashion. This effect was inhibited by a rat CRH antiserum that was administered to the rats 1 h before the TNF alpha injections. This suggested that CRH is a major mediator of the HPA axis response to TNF alpha. We subsequently evaluated the ability of TNF alpha to influence CRH and ACTH secretion in vitro by explanted rat hypothalami in organ culture and by dispersed rat anterior pituicytes in primary culture respectively. Hypothalami were incubated for 40 min with graded concentrations of TNF alpha (10 pM to 1 microM). This cytokine stimulated CRH secretion in a dose-dependent fashion, with an EC50 of 6.7 x 10 pM (P less than 0.05). Preincubation of hypothalamic explants with dexamethasone, indomethacin (1 microM), eicosatetraynoic acid (10 microM), or nordihydroguaiaretic acid (30 microM) resulted in inhibition of TNF alpha-stimulated CRH secretion (P less than 0.05). Interestingly, 4-h incubation with TNF alpha had no effect on ACTH secretion from rat anterior pituicytes at a concentration of 10 nM. Higher concentrations of TNF alpha (100 nM and 1 microM), however, elicited a dose-dependent increase in the ACTH concentration in the medium. Our results suggest that TNF alpha represents one of the immune response mediators that directly or via stimulation of other cytokines act as activators of the HPA axis during immune/inflammatory reactions. This effect appears to be glucocorticoid suppressible and eicosanoid mediated. The primary site of action of TNF alpha appears to by the hypothalamic CRH-secreting neuron. Some pituitary and adrenal effects of TNF alpha, however, cannot be excluded.     

Blunted adrenocorticotropin but normal beta-endorphin release after human corticotropin-releasing hormone administration in depression

Since the discovery of CRH in 1981, several investigators have reported abnormalities of the hypothalamic-pituitary-adrenal (HPA) system in response to direct stimulation of the corticotroph cells in patients with psychiatric disorders. To further explore HPA system integrity in major depressive disorders, 13 drug-free patients and normal subjects matched for age, sex, ovarian status, and body weight received 100 micrograms synthetic human CRH as an iv bolus dose. Compared to that in the normal subjects, in the depressed patients a significant attenuation of the net ACTH release after CRH administration (772 +/- 597 vs. 263 +/- 286 pmol/min.L; P less than 0.02) was observed, while beta-endorphin and cortisol responses did not differ significantly between the groups. The magnitudes of ACTH and cortisol release were negatively correlated in the patient group only (r = -0.67; P less than 0.01). Thus, the blunted ACTH response to CRH in depression might be related to hypercortisolemia, while the implications of the apparent dissociation of ACTH and beta-endorphin after CRH administration still remain unclear. Our data support the hypothesis that the hyperactivity of the HPA system in depression most likely is a consequence of CRH hypersecretion, the origin of which may be explained by abnormal central glucocorticoid receptor or neurotransmitter regulation.     

Hypothalamic somatostatin mediates the suppression of growth hormone secretion by centrally administered thyrotropin-releasing hormone in conscious rats

The effect and mechanism of action of central TRH on the regulation of GH secretion was studied in conscious male rats with indwelling intraatrial and intracerebroventricular (icv) cannulae. Plasma GH was measured every 10-20 min from 1000 h-1400 h by repeated blood sampling. In animals that received saline iv or icv, GH secretion was pulsatile, with peak hormone levels occurring at 1120-1200 h. TRH (10 micrograms), injected icv at 1100 h, inhibited spontaneous GH secretion, and mean plasma GH levels remained suppressed (less than 20 ng/ml) for at least 3 h after injection. In contrast, an iv injection of the same dose of TRH at 1100 h did not significantly affect spontaneous GH secretion. Intravenous injection of human GH-releasing factor [1-40] (hGRF, 1 micrograms) at 1100 h in animals injected 5 min earlier with saline (10 microliters, icv) stimulated GH release, with peak values (748 +/- 63 ng/ml, mean +/- SE) observed 10 min after injection. However, animals injected icv with TRH (10 micrograms) 5 min before the iv injection of hGRF exhibited an attenuated GH response to hGRF (peak values, 115 +/- 28 ng/ml; P less than 0.001 vs. saline icv + hGRF). The inhibition of GH secretion by central TRH was abolished by pretreatment of animals with antisomatostatin serum (0.5 ml, iv) but not with normal serum (P less than 0.001). These results suggest an inhibitory role of central TRH in the regulation of spontaneous GH secretion in the rat that is mediated by stimulation of hypothalamic somatostatin.     

Inhibition of adrenocorticotropin secretion by somatostatin in pituitary cells in culture

AtT20/D16v is a clonal strain of mouse pituitary tumor cells which synthesizes and secretes ACTH. Somatostatin, a hypothalamic tetradecapeptide, has been shown to inhibit the release of PRL, GH, and TSH from the pituitary gland. We have characterized specific binding sites for somatostatin on AtT20/D16v cells and demonstrate that somatostatin inhibits stimulated ACTH release by these cells. Equilibrium binding studies with [125I]Tyr1]somatostatin showed the presence of a single class of noninteracting binding sites on AtT20/D16v cells. Half-maximal binding of somatostatin occurred at 1.7 X 10(-9) M, and there were 26,300 binding sites/cell. The binding of [125I]Tyr1]somatostatin was not significantly inhibited by the hypothalamic peptides TRH, LHRH, and substance P. Somatostatin had no consistent effect on basal ACTH secretion by AtT20/D16v cells, but it inhibited ACTH secretion stimulated with either 50 mM KCl or a hypothalamic extract. Half-maximal inhibition occurred with 4 X 10(-10) M somatostatin. TRH, LHRH, and substance P at concentrations of 10(-7) M were without effect. Somatostatin had no effect on either basal or stimulated hormone secretion by GH12C1 or F4C1 cells, two cell strains which lack specific somatostatin-binding sites. A critical concentration of extracellular calcium was required for the stimulation of ACTH secretion in AtT20/D16v cells. No response to 50 mM KCl occurred in the presence of EGTA or cobalt. Increased extracellular calcium overcame the inhibition of stimulated hormone secretion by EGTA, cobalt, and somatostatin. Therefore, we conclude that the inhibition of stimulated ACTH secretion by somatostatin involves the interaction of the peptide with specific binding sites on AtT20/D16v cells and the inhibition of stimulus-elicited calcium influx.     

Suppression of ectopic adrenocorticotropin secretion by the long-acting somatostatin analog octreotide

The long-acting somatostatin analog (octreotide) was administered to a 37-yr-old woman with the ectopic ACTH syndrome. The patient had diffuse metastatic spread of a nonpituitary tumor, presumably of pancreatic origin, and severe and rapidly progressive hypercortisolism with extreme myopathy, hypokalemia, and diabetes mellitus. Plasma ACTH and lipotropin levels and 24-h urinary cortisol excretion were greatly elevated [218 pg/mL (48 pmol/L), 1340 pg/mL (220 pmol/L), and up to 830 micrograms/24 h (2290 nmol/day), respectively]. Urinary cortisol excretion decreased to normal within 3 days after the initiation of octreotide therapy (150, 300, and 600 micrograms/day), and plasma ACTH and lipotropin levels also decreased. Urinary cortisol excretion remained normal for 2 months during chronic octreotide therapy, and her general condition improved dramatically. The only side-effect was a slight increase in the number of bowel movements. Tumor progression, however, was not controlled, and she eventually died of hepatic insufficiency. These data indicate that octreotide can be a highly effective treatment for patients with the ectopic ACTH syndrome.     

Human plasma proopiomelanocortin N-terminal peptide and adrenocorticotropin: circadian rhythm, dexamethasone suppression, and corticotropin-releasing hormone stimulation

The circadian rhythm, suppression with dexamethasone, and stimulation by corticotropin-releasing hormone (CRH) of plasma immunoreactive (IR) proopiomelanocortin N-terminal (NT) and IR-ACTH were studied in nine normal subjects and two patients with Addison's disease. The RIA for human NT (hNT) used was specific for NT except for partial cross-reactivity with gamma 2MSH. In normal subjects, plasma IR-hNT and IR-ACTH had almost parallel circadian rhythms and were suppressed by dexamethasone. The mean plasma levels of IR-hNT and IR-ACTH at 0800 h were 140 +/- 23 (SD) and 23 +/- 5 pg/ml, respectively. Plasma IR-hNT increased in parallel with IR-ACTH 15 to 30 min after iv injection of 100 micrograms ovine CRH. Maximum percent increases in plasma IR-hNT and IR-ACTH were 185 +/- 47 and 235 +/- 10%, respectively. In Addison's disease, on the other hand, plasma levels of IR-hNT and IR-ACTH were markedly elevated and the circadian rhythms were parallel. The mean plasma IR-hNT and IR-ACTH levels at 0900 h were 4363 and 1750 pg/ml, respectively. These results suggest that plasma hNT and ACTH are produced from a common precursor in the pituitary gland and secreted concomitantly under various physiological conditions such as stimulation by CRH and inhibition by glucocorticoid.     

The corticotropin-releasing hormone test in normal short children: comparison of plasma adrenocorticotropin and cortisol responses to human corticotropin-releasing hormone and insulin-induced hypoglycemia

The human corticotropin-releasing hormone (hCRH) tests were performed in twelve normal short children, and the responses of plasma ACTH and cortisol to iv administration of 1 micrograms/kg hCRH were compared with those to insulin-induced hypoglycemia. After administration of hCRH, the mean plasma ACTH level rose from a basal value of 3.3 +/- 0.4 pmol/l (mean +/- SEM) to a peak value of 9.2 +/- 0.8 pmol/l at 30 min, and the mean plasma cortisol level rose from a basal value of 231 +/- 25 nmol/l to a peak value of 546 +/- 30 nmol/l at 30 min. The ACTH response after insulin-induced hypoglycemia was greater than that after hCRH administration; the mean peak level (P less than 0.01), the percent maximum increment (P less than 0.01), and the area under the ACTH response curve (P less than 0.01) were all significantly greater after insulin-induced hypoglycemia than those after hCRH administration. Although the mean peak cortisol level after insulin-induced hypoglycemia was about 1.3-fold higher than that after hCRH administration (P less than 0.01), neither the percent maximum increment in plasma cortisol nor the area under the cortisol response curve after insulin-induced hypoglycemia was significantly different from that after hCRH administration. Consequently, the acute increases in plasma ACTH after the administration of 1 microgram/kg hCRH stimulated the adrenal gland to almost the same cortisol response as that obtained with a much greater increase in plasma ACTH after insulin-induced hypoglycemia. These results suggest that a plasma ACTH peak of 9-11 pmol/l produces near maximum acute stimulation of adrenal steroidogenesis.     

Ectopic adrenocorticotropin syndrome caused by lung cancer that responded to corticotropin-releasing hormone

ACTH responses to corticotropin-releasing hormone (CRH) were studied in three patients with the ectopic ACTH syndrome caused by lung cancer. Plasma ACTH responded to synthetic CRH in two of three patients. Tumor tissues obtained from these two patients contained CRH and ACTH. In one patient, tumor ACTH secretion was stimulated by CRH in vitro. Tumor CRH was immunologically, chromatographically, and biologically similar to hypothalamic CRH. In addition, multiple forms of immunoreactive beta-endorphin were present in plasma and the tumor extracts. From these results, we conclude that some patients with the ectopic ACTH syndrome have tumors that produce both ACTH and CRH and that CRH can stimulate ACTH secretion by such tumors. Other patients with the ectopic ACTH syndrome do not have ACTH responses to CRH. Therefore, procedures other than CRH testing are needed to differentiate patients with Cushing's syndrome due to ectopic ACTH/CRH production from those with Cushing's disease, since the latter also usually have ACTH responses to CRH.     

Arginine stimulates growth hormone secretion by suppressing endogenous somatostatin secretion

To determine how arginine (Arg) stimulates GH secretion, we investigated its interaction with GHRH in vivo and in vitro. Six normal men were studied on four occasions: 1) Arg-TRH, 30 g arginine were administered in 500 mL saline in 30 min, followed by an injection of 200 micrograms TRH; 2) GHRH-Arg-TRH, 100 micrograms GHRH-(1-44) were given iv as a bolus immediately before the Arg infusion, followed by 200 micrograms TRH, iv; 3) GHRH test, 100 micrograms GHRH were given as an iv bolus; and 4) TRH test, 200 micrograms TRH were given iv as a bolus dose. Blood samples were collected at 15-min intervals for 30 min before and 120 min after the start of each infusion. Anterior pituitary cells from rats were coincubated with Arg (3, 6, 15, 30, and 60 mg/mL) and GHRH (0.05, 1, 5, and 10 nmol/L) for a period of 3 h. Rat GH was measured in the medium. After Arg-TRH the mean serum GH concentration increased significantly from 0.6 to 23.3 +/- 7.3 (+/- SE) micrograms/L at 60 min. TRH increased serum TSH and PRL significantly (maximum TSH, 11.1 +/- 1.8 mU/L; maximum PRL, 74.6 +/- 8.4 micrograms/L). After GHRH-Arg-TRH, the maximal serum GH level was significantly higher (72.7 +/- 13.4 micrograms/L) than that after Arg-TRH alone, whereas serum TSH and PRL increased to comparable levels (TSH, 10.2 +/- 3.0 mU/L; PRL, 64.4 +/- 13.6 micrograms/L). GHRH alone increased serum GH to 44.9 +/- 9.8 micrograms/L, significantly less than when GHRH, Arg, and TRH were given. TRH alone increased serum TSH to 6.6 +/- 0.6 mU/L, significantly less than the TSH response to Arg-TRH. The PRL increase after TRH only also was lower (47.2 +/- 6.8 micrograms/L) than the PRL response after Arg-TRH. In vitro Arg had no effect on basal and GHRH-stimulated GH secretion. Our results indicate that Arg administered with GHRH led to higher serum GH levels than did a maximally stimulatory dose of GHRH or Arg alone. The serum TSH response to Arg-TRH also was greater than that to TRH alone. We conclude that the stimulatory effects of Arg are mediated by suppression of endogenous somatostatin secretion.     

Involvement of corticotropin-releasing factor and somatostatin in stress-induced inhibition of growth hormone secretion in the rat

Corticotropin-releasing factor (CRF), which is released into the pituitary portal blood during exposure to noxious stimuli, can act centrally to inhibit GH secretion. We have investigated a possible role of endogenous CRF in mediating the stress-induced inhibition of GH release in the rat. While exposure to electroshocks markedly lowered plasma GH levels measured 20 min later, the central administration of the CRF antagonist alpha-Hel CRF-(9-41) totally abolished the effect of stress. To examine possible mechanisms through which CRF might mediate the inhibitory action of various stimuli on GH secretion, we have administered antisomatostatin (anti-SS) serum to CRF-injected rats and observed that immunoneutralization of endogenous SS blocked the inhibitory action of CRF on basal plasma GH values. Additionally, we have shown that CRF acted centrally to prevent the stimulatory action of both exogenously administered GH-releasing factor and the endogenous GH-releasing factor released by morphine sulfate. These effects were abolished by previous treatment with anti-SS serum. Such observations support the hypothesis that in the rat, endogenous CRF mediates the inhibitory action of noxious stimuli on GH secretion and further suggest that this effect may involve an increased release of endogenous SS.     

Interrelationship between the novel peptide ghrelin and somatostatin/growth hormone-releasing hormone in regulation of pulsatile growth hormone secretion

GH is an anabolic hormone that is essential for normal linear growth and has important metabolic effects throughout life. The ultradian rhythm of GH secretion is generated by the intricate patterned release of two hypothalamic hormones, somatostatin (SRIF) and GHRH, acting both at the level of the pituitary gland and within the central nervous system. The recent discovery of ghrelin, a novel GH-releasing peptide identified as the endogenous ligand for the GH secretagogue receptor and shown to induce a positive energy balance, suggests the existence of an additional neuroendocrine pathway for GH control. To further understand how ghrelin interacts with the classical GHRH/SRIF neuronal system in GH regulation, we used a combined physiological and histochemical approach. Our physiological studies of the effects of ghrelin on spontaneous pulsatile GH secretion in conscious, free-moving male rats demonstrate that 1) ghrelin, administered either systemically or centrally, exerts potent, time-dependent GH-releasing activity under physiological conditions; 2) ghrelin is a functional antagonist of SRIF, but its GH-releasing activity at the pituitary level is not dependent on inhibiting endogenous SRIF release; 3) SRIF antagonizes the action of ghrelin at the level of the pituitary gland; and 4) the GH response to ghrelin in vivo requires an intact endogenous GHRH system. Our dual chromogenic and autoradiographic in situ hybridization experiments provide anatomical evidence that ghrelin may directly modulate GHRH mRNA- and neuropeptide Y mRNA-containing neurons in the hypothalamic arcuate nucleus, but that SRIF mRNA-expressing cells are not major direct targets for ghrelin. Together, these findings support the idea that ghrelin may be a critical hormonal signal of nutritional status to the GH neuroendocrine axis serving to integrate energy balance and the growth process.