Twenty-four-hour pattern of growth hormone secretion in the rhesus monkey: studies including alterations of the sleep/wake and sleep stage cycles

Abstract

The 24-h pattern of GH secretion and its possible relation to the sleep/wake cycle and to sleep stages were studied in 12 male rhesus monkeys. Blood samples were drawn every 15 min for 96 h, 24 h, or shorter periods of time through chronic right atrial catheters which extended through the wall into the adjacent room. In addition, activity rating (daytime) and determination of sleep stages from electroencephalogram recordings (nighttime) were done. GH profiles were obtained during undisturbed conditions and during deprivation of nap, 5 h total sleep, slow wave sleep (SWS), and rapid eye movement (REM) sleep. GH secretion was episodic, with peak concentrations often exceeding 20 ngeq/ml and nadirs mostly below 1 ngeq/ml. Autocorrelation analysis demonstrated a circadian and an ultradian rhythm during undisturbed conditions. However, the cycle length of the ultradian rhythm showed large inter- and intraindividual variations (from 3--6 h). Neither cross-correlation analysis between hormonal and activity/electroencephalogram sleep stage time series nor results of deprivation experiments produced evidence for a link between nap phases, the sleep/wake cycle, or the SWS/REM sleep stage cycle on the one hand and the GH secretory pattern on the other hand. However, while SWS deprivation was highly effective, REM deprivation did not substantially reduce total REM sleep time due to frequent entries into abortive REM sleep epochs. During the daytime, there was no significant correlation between activity/arousal and GH, but during the night, there was a significant positive correlation between stage waking and GH. A direct or indirect synchronizing effect of the matutinal light change is suggested by the pattern of the 24-h curve of mean GH concentrations during undisturbed conditions: a steep increase from very low concentrations at light onset, followed by a succession of nadirs and peaks at approximately 4.5-h intervals. However, the nadirs became progressively more shallow until there was no apparent periodicity during the night due to the loss of synchronization. It is concluded that GH in the rhesus monkey shows a circadian and an ultradian periodicity. However, in contrast to man, sleep and SWS are not important determinators of the 24-h GH pattern.