Methoscopolamine inhibition of sleep-related growth hormone secretion. Evidence for a cholinergic secretory mechanism.

We have examined the effects of cholinergic blockade with 0.5 mg methscopolamine bromide, intramuscularly, on sleep-related and insulin-induced growth hormone (GH) secretion. 17 normal young men were studied; 8 had sleep studies, and 12 (including 3 who also had sleep studies) had insulin tolerance tests (ITT) with 0.1 U/kg of regular insulin. After an adjustment night in the sleep laboratory, saline control night and methscopolamine night studies were done in random sequence; study procedures included electroencephalographic, electromyographic, and electrooculographic recordings, and blood sampling every 20 min for hormone radioimmunoassays. Prolactin levels were also measured during sleep. For methscopolamine night studies, the mean overall control GH level of 2.89+/-0.44 ng/ml and the mean peak control GH level of 11.09+/-3.11 ng/ml were dramatically reduced to 0.75+/-0.01 and 1.04+/-0.25 ng/ml, respectively (P<0.0001 and <0.001). Despite virtual absence of GH secretion during the night in every study subject, no measured sleep characteristic was affected by methscopolamine, including total slow-wave sleep (12.1+/-2.6% control vs. 10.3+/-2.5% drug, P>0.2). Sleep prolactin levels were not changed by methscopolamine. In contrast to the abolition of sleep-related GH secretion, administration of methscopolamine had only a marginal effect on the GH response to insulin hypoglycemia. None of nine time points differed significantly, as was also the case with peak levels, mean increments, and areas under the curves (P>0.2). Analysis of variance did, however, indicate that the lower GH concentrations achieved during ITT after methscopolamine (average 31.7% below control) were significantly different than control concentrations. We conclude that the burst of GH secretion which normally occurs after sleep onset is primed by a cholinergic mechanism which does not influence slow-wave sleep. Cholinergic mechanisms do not appear to play an important role in sleep-related prolactin secretion. The contrast between the complete suppression of sleep-related GH release and the relatively small inhibitory effect on ITT-induced GH secretion suggests that the neurotransmitter mechanisms, and presumably the pathways, which subserve sleep-related GH secretion in man may be different from those which mediate the GH response to pharmacologic stimuli such as insulin.