Tripartite neuroendocrine activation of the human growth hormone (GH) axis in women by continuous 24-hour GH-releasing peptide infusion: pulsatile, entropic, and nyctohemeral mechanisms.

Despite the discovery of potent GH-releasing peptides (GHRPs) more than 15 yr ago and the recent cloning of human, rat, and pig GHRP receptors in the hypothalamus and pituitary gland, the neuroregulatory mechanisms of action of GHRP agonists on the human hypothalamo-somatotroph unit are not well delineated. To gain such clinical insights, we evaluated the ultradian (pulsatile), entropic (pattern orderliness), and nyctohemeral GH secretory responses during continuous 24-h i.v. infusion of saline vs. the most potent clinically available hexapeptide, GHRP-2 (1 microg/kg x h) in estrogen-unreplaced (mean serum estradiol, 12 +/- 2.4 pg/mL) postmenopausal women (n = 7) in a paired, randomized design. Blood was sampled every 10 min for 24 h during infusions and was assayed by ultrasensitive GH chemiluminescence assay. Pulsatile GH secretion was quantitated by deconvolution analysis, orderliness of GH release patterns by the approximate entropy statistic, and 24-h GH rhythmicity by cosinor analysis. Statistical analysis revealed that GHRP-2 elicited a 7.7-fold increase in (24-h) mean serum (+/-SEM) GH concentrations, viz. from 0.32 +/- 0.042 (saline) to 2.4 +/- 0.34 microg/L (GHRP-2; P = 0.0006). This occurred via markedly stimulated pulsatile GH release, namely a 7.1-fold augmentation of GH secretory burst mass: 0.87 +/- 0.18 (control) vs. 6.3 +/- 1.3 microg/L (GHRP-2; P = 0.0038). Enhanced GH pulse mass reflected a commensurate 10-fold (P = 0.023) rise in GH secretory burst amplitude [maximal GH secretory rate (micrograms per L/min) attained within a secretory pulse] with no prolongation in event duration. GH burst frequency, interpulse interval, and calculated GH half-life were all invariant of GHRP-2 treatment. Concurrently, as detected in the ultrasensitive GH assay, GHRP-2 augmented deconvolution-estimated interpulse (basal) GH secretion by 4.5-fold (P = 0.025). The approximate entropy of 24-h serum GH concentration profiles rose significantly during GHRP-2 infusion; i.e. from 0.592 +/- 0.073 (saline) to 0.824 +/- 0.074 (GHRP-2; P = 0.0011), signifying more irregular or disorderly GH release patterns during secretagogue stimulation. Cosinor analysis of 24-h GH rhythms disclosed a significantly earlier (daytime) acrophase at 2138 h (+/- 140 min) during GHRP-2 stimulation vs. 0457 h (+/-42 min) during saline infusion (P = 0.013). Concomitantly, the cosinor amplitude rose 6-fold (P = 0.018), and the mesor (cosine mean) rose 5-fold (P = 0.003). Fasting (0800 h) plasma insulin-like growth factor (IGF-I) concentrations rose by -11 +/- 12 microg/L during saline infusion and by 102 +/- 18 microg/L during GHRP-2 infusion (P = 0.0036). GHRP-2 infusion did not modify (24-h pooled) serum LH, FSH, or TSH concentrations and minimally increased serum (pooled) daily PRL (6.8 +/- 0.83 vs. 12 +/- 1.2 microg/L; P < 0.05) and cortisol (5.3 +/- 0.59 to 7.0 +/- 0.74; P < 0.05) concentrations. In summary, 24-h constant iv GHRP-2 infusion in the gonadoprival female neurophysiologically activates the GH-IGF-I axis by potentiating GH secretory burst mass and amplitude by 7- to 10-fold and augmenting the basal (nonpulsatile) GH secretion by 4.5-fold. GHRP-2 action is highly selective, as it does not alter GH secretory burst frequency, interpulse interval, event duration, or GH half-life. GHRP-2 effectively elevates IGF-I concentrations, unleashes greater disorderliness of GH release patterns, and heightens the 24-h rhythmicity of GH secretion. These tripartite features of GHRP-2's action in estrogen-withdrawn (postmenopausal) women also characterize normal human puberty and/or sex steroid regulation of the GH-IGF-I axis. However, how or whether GHRP-2 interacts further with sex hormone modulation of GH neurosecretory control in older women and men is not yet known.