Whether the biological effect of melatonin in humans is directly relat
ed to the circulating levels of the hormone, has not heretofore been i
nvestigated. In this study, we investigated whether previously describ
ed hypothermic melatonin properties are dose related. The nocturnal de
cline of the body temperature (BT) observed in 16 early follicular pha
se women, following placebo administration at 18.00 h, was compared wi
th that observed during the preceding or following night, after melato
nin suppression with the beta(1)-adrenergic antagonist atenolol (100 m
g). In 6 subjects (37.5%) with lower nocturnal melatonin levels (p < 0
.05) atenolol induced a complete melatonin suppression and an attenuat
ion of the nocturnal BT decline (p < 0.02), whereas in the remaining 1
0 subjects (62.5%) atenolol induced an incomplete melatonin suppressio
n with no modification of the nocturnal BT decline. During a 3rd night
, 2 of the 6 subjects with complete and 6 of the 10 subjects with inco
mplete melatonin suppression blindly received atenolol plus melatonin
(1 mg at 19.30h and 0.75 mg at 21.00 and 23.00 h). Exogenous melatonin
restored the full expression of the nocturnal BT decline in the 2 sub
jects with complete melatonin suppression, but did not modify the BT d
ecline in the 6 subjects with atenolol-induced incomplete melatonin su
ppression. Our data show that markedly, but not completely attenuated
nocturnal melatonin levels are sufficient to exert maximal thermoregul
atory effects, indicating rather a threshold than a dose-response effe
ct of melatonin action on human BT.