Pulsatile growth hormone secretion persists in genetic growth hormone-releasing hormone resistance

Growth hormone (GH) secretion is regulated by GH-releasing hormone (GHRH), somatostatin, and possibly ghrelin, but uncertainty remains about the relative contributions of these hypophysiotropic factors to GH pulsatility. Patients with genetic GHRH receptor (GHRH-R) deficiency present an opportunity to examine GH secretory dynamics in the selective absence of GHRH input. We studied circadian GH profiles in four young men homozygous for a null mutation in the GHRH-R gene by use of an ultrasensitive GH assay. Residual GH secretion was pulsatile, with normal pulse frequency, but severely reduced amplitude (<1% normal) and greater than normal process disorder (as assessed by approximate entropy). Nocturnal GH secretion, both basal and pulsatile, was enhanced compared with daytime. We conclude that rhythmic GH secretion persists in an amplitude-miniaturized version in the absence of a GHRH-R signal. The nocturnal enhancement of GH secretion is likely mediated by decreased somatostatin tone. Pulsatility of residual GH secretion may be caused by oscillations in somatostatin and/or ghrelin; it may also reflect intrinsic oscillations in somatotropes.     

Growth hormone regulation of growth hormone-releasing hormone gene expression

Slot-blot hybridization technique was used to evaluate growth hormone-releasing hormone (GHRH) mRNA levels in the hypothalamus of long-term (14 days) hypophysectomized (HPX) rats treated or not with 125 micrograms hGH/rat, twice daily IP, since the first day postsurgery. In addition, mRNA levels were determined in the hypothalamus of short-term (4 days) GH-treated (250 micrograms hGH/rat, twice daily IP) intact rats. GHRH mRNA levels were increased in HPX rats, and GH treatment partially counteracted this rise. Short-term administration of GH decreased GHRH mRNA levels in intact rats. These results, evaluated together with previous findings showing decreased hypothalamic GHRH-like immunoreactivity in both HPX rats and intact rats given GH (6, 7, 9), indicate that GH exerts a negative feedback action on the synthesis and release of GHRH.     

Regulation of growth hormone-releasing hormone gene expression and biosynthesis

Growth hormone-releasing hormone (GRH) was initially isolated, characterized, sequenced, and cloned from human tumors and subsequently from the hypothalamus of humans and other animal species. Extensive structure-function studies have indicated the amino terminus to be most important for its biologic action, and the primary mechanism of its bioinactivation occurs by cleavage of an amino terminal dipeptide. The GRH gene is expressed primarily in the hypothalamic arcuate nucleus but also in the placenta. Expression of the GRH gene is regulated by growth hormone in a classical feedback manner, with hypophysectomy leading to increased expression that is reversed by growth hormone treatment. GRH gene overexpression in transgenic mice leads to a syndrome similar to that of ectopic GRH secretion with massive pituitary hyperplasia and markedly enhanced growth. The transgenic mouse has been used for studies of GRH biosynthesis and provides a suitable model for the study of precursor processing to the mature hormone.     

Factors influencing the growth hormone response to growth hormone-releasing hormone in children with idiopathic growth hormone deficiency

OBJECTIVE: To evaluate the factors influencing the growth hormone (GH) response to GH-releasing hormone (GHRH) test in idiopathic GH deficiency. METHODS: 28 patients aged 4.9 +/- 0.7 years with certain GH deficiency were given GHRH (2 microg/kg). RESULTS: The GH peak after GHRH was correlated negatively with age at evaluation (r = -0.37, p < 0.05) and body mass index (r = -0.44, p = 0.02), and positively with anterior pituitary height (r = 0.47, p = 0.02), GH peak after non-GHRH stimulation (r = 0.78, p < 0.0001) and spontaneous GH peak (r = 0.82, p = 0.007). It was lower in the patients aged >5 years than in the youngest (p = 0.04), but it was similar in the patients with and without features suggesting a hypothalamic origin. CONCLUSION: The GH response to GHRH test cannot be used to differentiate between hypothalamic and pituitary forms of idiopathic GH deficiency, probably because the GH response decreases after the first 5 years of life, whatever the origin of the deficiency.     

Plasma growth hormone secretion during constant growth hormone-releasing factor infusion

Synthetic human pancreatic growth hormone-releasing factor (hpGRF-44) was infused intravenously at a constant rate of 2.5 micrograms/min for 180 minutes in 3 normal boys of short stature. Plasma GH levels reached a peak at 60-120 min with a mean value (+/- SEM) of 69.1 +/- 14.3 ng/ml, and then, declined gradually in spite of continuous hpGRF-44 infusion up to 180 minutes. Similarly, constant infusion of hpGRF-44 at a rate of 2.5 micrograms/min in 5 normal but short boys for 90 minutes, together with an iv bolus injection of hpGRF-44 (2 micrograms/kg) administered at 0 and 90 minutes, elicited a prompt rise in plasma GH 15-30 minutes after the first bolus but no significant elevation of GH was observed after the second bolus. In contrast, when two iv bolus injections of hpGRF-44 (2 micrograms/kg) were given in 4 normal boys with short stature at 0 and 90 minutes, respectively, significant elevation of plasma GH was found after each bolus. These results suggest that under constant infusion of GRF the pituitary experiences a down-regulation after the initial peak of GH response, possibly due to desensitization to GRF.     

Effect of growth hormone-releasing hormone and clonidine on growth hormone release in type 1 diabetic patients

We administered growth-hormone releasing hormone (GHRH), clonidine or thyrotropin-releasing hormone (TRH) as intravenous boli each in three different randomized mornings to nine well-controlled Type 1 diabetic men and to six age-matched healthy men who served as controls. GHRH and clonidine evoked a prompt and brisk GH release both in diabetic and in control subjects with no significant difference being evident between the two groups. Only one diabetic subject showed a paradoxical GH release after TRH when he was under long-term poor metabolic control. These results indicate that in insulin-dependent patients with good control of the metabolic disease the response of somatotropes to pituitary- or central nervous system-directed stimuli is normal. These data are supportive of the idea that altered GH secretion in Type 1 diabetes rather than reflecting a primary hypothalamic and/or pituitary alteration may be a state-dependent phenomenon related to the metabolic state of the disease.     

Feeding reduces activity of growth hormone-releasing hormone and somatostatin neurons

Secretion of growth hormone (GH) is synchronized among castrate male cattle (steers) around feeding when access to feed is restricted to a 2-hr period each day. Typically, concentrations of GH increase before and decrease after feeding. Our objectives were to determine whether i) concentrations of GH decrease in blood after start of feeding; ii) activity of immunoreactive growth hormone-releasing hormone (GHRH-ir) neurons decreases in the arcuate nucleus (ARC) after feeding; iii) activity of immunoreactive somatostatin (SS-ir) neurons in the periventricular nucleus (PeVN) and ARC increase after feeding; and iv) GHRH stimulates release of GH to a similar magnitude at 0900 and at 1300 hr, in steers fed between 1000 and 1200 hr. Blood samples were collected at 20-min intervals from 0700 to 1300 hr. Groups of steers were euthanized at 0700, 0900, 1100, and 1300 hr (n = 5 per group). Dual-label immunohistochemistry was performed on free-floating sections of hypothalami using antibodies directed against Fos and Fos-related antigens (Fos/FRA) as a marker of neuronal activity in immunoreactive GHRH and SS neurons. Concentrations of GH were high before and decreased after feeding. The percentage of SS-ir neurons containing Fos/FRA-ir in the PeVN was 50% lower (P<0.01) at 1100 hr and 36% lower (P<0.05) at 1300 hr than at 0900 hr. There was no change in percentage of SS-ir neurons containing Fos/FRA-ir in the ARC. The percentage of GHRH-ir neurons containing Fos/FRA-ir in the ARC was 66% lower (P<0.05) at 1100 hr and 65% lower (P<0.05) at 1300 hr than at 0700 hr. In contrast, the number of GHRH-ir neurons increased from 0700 to 1300 hr. GHRH-induced release of GH was suppressed at 1300 hr compared with 0900 hr. In conclusion, reduced basal and GHRH-induced secretion of GH after feeding was associated with decreased activity of GHRH neurons in the ARC and decreased activity of SS neurons in the PeVN.     

Interactions between growth hormone-releasing hormone and glucocorticoids in male rats

While chronic glucocorticoid treatment increases pituitary growth hormone (GH) content in rats and primates and increases pituitary GH release in response to growth hormone-releasing hormone (GHRH) in rats, it also inhibits somatic growth. We investigated these opposite actions in rats using the synthetic glucocorticoid dexamethasone. Seven days of dexamethasone treatment (40 micrograms/animal per day) did not alter the frequency of spontaneous GH pulses in conscious, freely-moving animals. The amplitude of the GH pulses in saline and dexamethasone-treated rats was different (P less than 0.01), the latter group having a higher incidence of GH levels less than 95 ng/ml, a lower incidence of GH levels between 96 and 251 ng/ml, and a higher incidence of GH values greater than 480 ng/ml. A 20 microgram/kg per day dose of dexamethasone was sufficient to significantly inhibit growth but was inadequate in enhancing the GH response to an acute injection of GHRH in anesthetized animals. These results support the concept that glucocorticoids exert their catabolic effects on somatic growth in peripheral tissues and not at the pituitary level.     

Sensitivity of rat forebrain neurons to growth hormone-releasing hormone

The effects of iontophoretically applied human pancreatic growth hormone-releasing factor (hpGRF), peptide histidine isoleucine (PHI-27), and somatostatin (SS) on the extracellular activity of single cells in the hypothalamus, thalamus, and cortex of the rat brain were studied in urethane-anesthetized, male rats. Neurons with membrane sensitivity to hpGRF, PHI-27, and SS were present in each brain region. Although neurons excited by these peptides were encountered in thalamus and hypothalamus, depression of neuronal firing was the predominant response observed. Overall, the neurons responding to hpGRF also possessed membrane sensitivity to PHI-27, whereas, the hpGRF sensitive neurons appeared to be more divided as to their ability to respond to SS. The results clearly demonstrate that hpGRF and PHI-27 are capable of affecting the membrane excitability of neurons in several brain regions. The distribution of neurons sensitive to hpGRF suggests that hypothalamic GRF, in addition to its well documented role in the regulation of pituitary growth hormone secretion, may subserve other physiological events in the rat central nervous system as a neurotransmitter and/or neuromodulator.     

Role of aging on growth hormone and prolactin release after growth hormone-releasing hormone and domperidone in man

Growth hormone (GH) and prolactin (PRL) secretion after GH-releasing hormone (GHRH) and domperidone (DOM), an antidopaminergic drug which does not cross the blood-brain barrier (BBB), was evaluated in 8 healthy elderly men (65-91 years) and in 7 young adults (23-40 years). All received in random order at 2-day intervals: GHRH(1-40) (50 micrograms i.v.) bolus, DOM (5 mg/h) infusion, GHRH(1-40) (50 micrograms i.v.) plus DOM (5 mg/h i.v.), saline solution. In elderly men GH increase after GHRH was significantly lower than in young men. DOM alone did not change GH secretion in either of these groups, whereas it increased the GH response to GHRH only in young adults. PRL levels increased in both young and elderly men during both DOM and GHRH plus DOM, but the PRL release was more marked in young than in elderly men. Both integrated secretion of GH after GHRH and of PRL after DOM were inversely correlated to chronological age. Our data show an impairment of GH rise after GHRH and of PRL after DOM in elderly adults. It is also stressed that peripheral blockade of dopamine receptors by DOM is unable to amplify the GH response to GHRH only in elderly men. A reduction in GH release after GHRH might be related to aging, perhaps through a reduction of dopaminergic tonus.     

Subcutaneous treatment with growth hormone-releasing hormone for short stature

In the present study we report the effects of therapy with growth hormone-releasing factor (1-29)NH2 (GRF) on growth rate, plasma levels of insulin growth factor I (IGF-I) and growth hormone (GH) secretion in 11 children who were selected solely on the basis of their short stature and normal GH secretion on standard provocative tests. All children received GRF for 6 months (5 micrograms/kg body weight subcutaneously) each evening. The 24-hour GH secretory profile was studied before and after 6 months of treatment. Simultaneously, GH secretory responses to single intravenous bolus GRF (1.5 micrograms/kg body weight) were also studied before, during, and 6 months off therapy with GRF(1-29)NH2. Plasma levels of IGF-I were measured before, during (1, 2 and 6 months), and after 6 months off therapy with GRF. Statural growth was measured at 3-month intervals. The peak plasma GH level in response to GRF was 56.04 +/- (SD) 24.46 ng/ml before treatment, and similar results were found after therapy. The 24-hour GH secretory profile did not show differences before, during, and after treatment. Comparably, no differences were found in GH pulse frequency, pulse amplitude, pulse height, pulse increment, pulse area and total area before, and 6 months off therapy with GRF. The increments in serum IGF-I achieved were not significantly different at all intervals studied. All patients increased growth velocities (mean +/- SD, cm/year) in response to GRF therapy. Our results demonstrate that GRF administration was effective in accelerating growth velocity in 11 children without GH deficiency.     

Sleep in mice with nonfunctional growth hormone-releasing hormone receptors

The role of the somatotropic axis in sleep regulation was studied by using the lit/lit mouse with nonfunctional growth hormone (GH)-releasing hormone (GHRH) receptors (GHRH-Rs) and control heterozygous C57BL/6J mice, which have a normal phenotype. During the light period, the lit/lit mice displayed significantly less spontaneous rapid eye movement sleep (REMS) and non-REMS (NREMS) than the controls. Intraperitoneal injection of GHRH (50 microg/kg) failed to promote sleep in the lit/lit mice, whereas it enhanced NREMS in the heterozygous mice. Subcutaneous infusion of GH replacement stimulated weight gain, increased the concentration of plasma insulin-like growth factor-1 (IGF-1), and normalized REMS, but failed to restore normal NREMS in the lit/lit mice. The NREMS response to a 4-h sleep deprivation was attenuated in the lit/lit mice. In control mice, intraperitoneal injection of ghrelin (400 microg/kg) elicited GH secretion and promoted NREMS, and intraperitoneal administration of the somatostatin analog octretotide (Oct, 200 microg/kg) inhibited sleep. In contrast, these responses were missing in the lit/lit mice. The results suggest that GH promotes REMS whereas GHRH stimulates NREMS via central GHRH-Rs and that GHRH is involved in the mediation of the sleep effects of ghrelin and somatostatin.     

Growth hormone-releasing hormone antagonists inhibit growth of human ovarian cancer

Epithelial ovarian carcinoma is the leading cause of cancer-related deaths among women with gynecologic malignancies. Antagonists of the growth hormone-releasing hormone (GHRH) have been shown to inhibit growth of various cancers through endocrine, autocrine, and paracrine mechanisms. In this study, we have investigated the effects of GHRH antagonists (GHRHa) in ES-2 human clear cell ovarian cancer and in UCI-107 human serous ovarian cancer in vitro and in vivo. We evaluated the expression of mRNA for GHRH receptor, the binding to GHRH receptors, in specimens of ES-2 ovarian cancer. We evaluated also the in vitro effects of GHRHa on ES-2 cells and the in vivo effect of 2 different GHRHa on ES-2 and UCI-107 tumors. Nude mice bearing xenografts on ES-2 and UCI-107 ovarian cancer were treated with JMR-132 and MZ-J-7-118, respectively. Tumor growth was compared to control. ES-2 cells expressed mRNA for the functional splice variant SV1 of the GHRH receptor. JMR-132 inhibited cell proliferation in vitro by 42% and 18% at 10 and 1 μM concentration, respectively. Specific high affinity receptors for GHRH were detected in ES-2 cancer samples. In vivo daily subcutaneous injections of GHRHa significantly reduced tumor growth compared to a control group in both animal models. Our results indicate that GHRHa such as JMR-132 and MZ-J-7-118 can inhibit the growth of human ovarian cancer. The efficacy of GHRHa in ovarian cancer should be assessed in clinical trials.     

Internalization and trafficking of the human and rat growth hormone-releasing hormone receptor

Internalization and intracellular trafficking of the growth hormone-releasing hormone receptor (GHRH-R) were studied in rat anterior pituitary and human (h) and rat (r) GHRH-R-transfected BHK cells, with the GHRH agonist, [N(alpha)-5-carboxyfluoresceinyl-D-Ala(2), Ala(8), Ala(15), Lys(22)]hGHRH(1-29)NH(2) (Fluo-GHRH). Time- and temperature-dependent internalization of stimulated GHRH-R was blocked by phenyl arsine oxide (PAO) in both cell types. In anterior pituitary and rGHRH-R-transfected BHK cells, only filipin III and cerulenin blocked receptor-mediated internalization of Fluo-GHRH while in hGHRH-R-transfected BHK cells, only hyperosmolar sucrose inhibited this process. These results suggest that hGHRH-R internalization is clathrin-dependent, while fatty acid acylation of rGHRH-R appears to be a prerequisite to caveolin-dependent internalization. Experiments in anterior pituitary using Bodipy-FL-C(5) ganglioside GM1, a specific marker of lipid rafts such as caveolae, confirmed this latter pathway. Co-localization of Fluo-GHRH with LysoTracker indicated that Fluo-GHRH was directed to acidic organelles in both cell types. Finally, studies using cycloheximide and monensin showed that upon stimulation with GHRH, an optimal concentration of functional GHRH-R was maintained at the plasma membrane due to de novo synthesis and recycling in pituitary cells and to de novo synthesis solely in hGHRH-R-transfected BHK cells. This first study on the dynamics of the GHRH/GHRH-R complexes using fluorescence imaging in a native environment compared to cell system models, revealed that both receptor primary structure and concentration at the plasma membrane play important roles in internalization and trafficking of specific G-protein-coupled receptors (GPCR).     

Effects of plasmid growth hormone-releasing hormone treatment during heat stress

A gene therapy treatment with plasmid-based growth hormone-releasing hormone (GHRH) delivered by electroporation (EP) was investigated during heat stress; 32 primiparous cows received 2.5 mg of a GHRH-expressing myogenic plasmid (pSP-HV-GHRH), while 20 were designated as controls. Offspring of treated animals showed a reduction in mortality (47%; p < 0.02), and survival from birth to 260 days was dramatically improved (0% mortality vs. 21% in controls) along with an increase in weight gain (p < 0.05). Milk production was increased compared to controls with an average yield gain of 421 kg/cow (p = 0.028). Prolactin (PRL) levels were also significantly increased compared to controls (p < 0.05). The second pregnancy rate was improved by GHRH treatment (53.3% vs. 30.8%). This study shows that the use of plasmid-mediated therapy delivered by EP can maintain health status during periods of heat stress, important for both animals and potentially humans in hot, challenging climates.     

Growth hormone response of bull calves to growth hormone-releasing factor

Three experiments were conducted to determine serum growth hormone (GH) response of bull calves (N = 4; 83 kg body wt) to iv injections and infusions of human pancreatic GH-releasing factor 1-40-OH (hpGRF). Peak GH responses to 0, 2.5, 10, and 40 micrograms hpGRF/100 kg body wt were 7 +/- 3, 8 +/- 3, 18 +/- 7, and 107 +/- 55 (mean peak height +/- SEM) ng/ml serum, respectively. Only the response to the 40-microgram dose was greater (P less than 0.05) than the 0-microgram dose. Concentrations of prolactin in serum were not affected by hpGRF treatment. In calves injected with hpGRF (20 micrograms/100 kg body wt) at 6-hr intervals for 48 hr, GH increased from a mean preinjection value of 3.1 ng/ml serum to a mean peak response value of 70 ng/ml serum. Differences in peak GH response between times of injection existed within individual calves (e.g., 10.5 ng/ml vs 184.5 ng/ml serum). Concentrations of GH in calves infused continuously with either 0 or 200 micrograms hpGRF/hr for 6 hr averaged 7.4 +/- 3 and 36.5 +/- 11 ng/ml serum, respectively (P less than 0.05). Concentrations of GH oscillated markedly in hpGRF-infused calves, but oscillations were asynchronous among calves. We conclude that GH response of bull calves to hpGRF is dose dependent and that repeated injections or continuous infusions of hpGRF elicit GH release, although magnitude of response varies considerably. We hypothesize that differences in GH response to hpGRF within and among calves, and pulsatile secretion in the face of hpGRF infusion may be related to the degree of synchrony among exogenous hpGRF and endogenous GRF and somatostatin.     

Biological activities of two porcine growth hormone-releasing hormone receptor isoforms

Binding of growth hormone-releasing hormone (GHRH) to two isoforms (G3R and G5R) of the porcine GHRH receptor was studied. Both G3R- and G5R-cDNA were isolated from a porcine anterior pituitary cDNA library and have an identical primary structure from aa 1 to 418 and a different aa sequence from aa 419 to 423. In addition, the G5R isoform contains an extra C-terminal tail of 28 aa. The G3R and G5R mRNAs arise from alternative splicing of a single precursor mRNA for GHRH receptors. A mammalian cell expression vector containing either G3R or G5R cDNA under the regulation of a strong human cytomegalovirus promoter was constructed and used to transfect a human embryonic kidney 293 cell line. Two stable transfectants (293/G3R-4 and 293/G5R-12) were isolated on the basis of high expression of the receptor mRNAs. Both G3R and G5R mRNAs were expressed at similarly high levels in 293/G3R-4 and 293/G5R-12 cells; however, GHRH binding to 293/G3R-4 cells was much greater than that observed for 293/G5R-12 cells. Basal as well as GHRH-stimulated GTPase activity and intracellular cAMP concentration are also significantly greater in 293/G3R-4 cells as compared to 293/G5R-12 cells. We conclude that the modification of GHRH receptor at the C-terminal region hindered GHRH binding to the receptor and thus attenuates its biological activities.     

Potent agonists of growth hormone-releasing hormone. Part I.

Analogs of the 29 amino acid sequence of growth hormone-releasing hormone (GH-RH) with agmatine (Agm) in position 29 have been synthesized by the solid phase method, purified, and tested in vitro and in vivo. The majority of the analogs contained desaminotyrosine (Dat) in position 1, but a few of them had Tyr1, or N-MeTyr1. Some peptides contained one or more additional L- or D-amino acid substitutions in positions 2, 12, 15, 21, 27, and/or 28. Compared to the natural sequence of GH-RH(1-29)NH2, [Dat1,Ala15]GH-RH(1-28)Agm (MZ-3-191) and [D-Ala2,Ala15]GH-RH(1-28)Agm (MZ-3-201) were 8.2 and 7.1 times more potent in vitro, respectively. These two peptides contained Met27. Their Nle27 analogs, [Dat1,Ala15,Nle27]GH-RH(1-28)Agm(MZ-2-51), prepared previously (9), and [D-Ala2,Ala15,Nle28]GH-RH(1-28)Agm(MZ-3-195) showed relative in vitro potencies of 10.5 and 2.4, respectively. These data indicate that replacement of Met27 by Nle27 enhanced the GH-releasing activity of the analog when the molecule contained Dat1-Ala2 residues at the N-terminus, but peptides containing Tyr1-D-Ala2 in addition to Nle27 showed decreased potencies. Replacement of Ser28 with Asp in multi-substituted analogs of GH-RH(1-28)Agm resulted in a decrease in in vitro potencies compared to the parent compound. Thus, the Ser28-containing MZ-2-51, and [Dat1,Ala15,D-Lys21,Nle27]GH-RH(1-28)Agm, its Asp28 homolog (MZ-3-149), possessed relative activities of 10.5 and 5.6, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)     

Growth hormone-releasing factor on growth hormone secretion in prepubertal calves

The effects of iv administration of growth hormone-releasing factor (GRF) on growth hormone (GH) release and on nitrogen metabolism were measured in prepubertal calves. Crossbred beef heifers (111 kg) were used in a Latin square design to test the effects of 0, 0.01, 0.033, 0.067, and 0.1 microgram human pancreatic (hp) GRF [hpGRF (1,40)OH]/kg body wt on plasma GH concentrations. When they were given doses of 0.067 and 0.1 microgram hpGRF/kg body wt, plasma GH increased (P less than 0.05) within 5-15 min, compared with injections of control buffer, and then returned to preinjection concentrations. The response to 0.067 microgram hpGRF/kg body wt every 3 hr for 42 hr was studied in five heifers (137 kg body wt). The animals responded to 50% of the GRF injections with an increase in plasma GH during every 6-hr period measured. Nitrogen retention, hormone concentrations, and weight gain were measured in five bull calves (90 kg body wt) administered 0 or 0.067 microgram Nle rat hypothalamic GRF (1,29)NH2/kg body wt every 4 hr for 10 days. Metabolic parameters were interpreted to indicate an anabolic response to GRF even though increases of 16% in nitrogen retention, 23% in plasma somatomedin C concentrations, and 36% in weight gain with pulsatile GRF treatment were variable and statistically similar to those of controls. These results indicate that GRF induces peak GH secretion within 15 min in prepubertal calves and that calves can respond to multiple injections of GRF with an increase in plasma GH.