Serotonin and the mammalian circadian system: I. In vitro phase shifts by serotonergic agonists and antagonists.

The primary mammalian circadian clock, located in the suprachiasmatic nuclei (SCN), receives a major input from the raphe nuclei. The role of this input is largely unknown, and is the focus of this research. The SCN clock survives in vitro, where it produces a 24-hr rhythm in spontaneous neuronal activity that is sustained for at least three cycles. The sensitivity of the SCN clock to drugs can therefore be tested in vitro by determining whether various compounds alter the phase of this rhythm. We have previously shown that the nonspecific serotonin (5-HT) agonist quipazine resets the SCN clock in vitro, inducing phase advances in the daytime and phase delays at night. These results suggest that the 5-HT-ergic input from the raphe nuclei can modulate the phase of the SCN circadian clock. In this study we began by using autoradiography to determine that the SCN contain abundant 5-HT1A and 5-HT1B receptors, very few 5-HT1C and 5-HT2 receptors, and no 5-HT3 receptors. Next we investigated the ability of 5-HT-ergic agonists and antagonists to reset the clock in vitro, in order to determine what type or types of 5-HT receptor(s) are functionally linked to the SCN clock. We began by providing further evidence of 5-HT-ergic effects in the SCN. We found that 5-HT mimicked the effects of quipazine, whereas the nonspecific 5-HT antagonist metergoline blocked these effects, in both the day and night. Next we found that the 5-HT1A agonist 8-OH-DPAT, and to a lesser extent the 5-HT1A-1B agonist RU 24969, mimicked the effects of quipazine during the subjective daytime, whereas the 5-HT1A antagonist NAN-190 blocked quipazine's effects. None of the other specific agonists or antagonists we tried induced similar effects. This suggests that quipazine acts on 5-HT1A receptors in the daytime to advance the SCN clock. None of the specific agents we tried were able either to mimic or to block the actions of 5-HT or quipazine at circadian time 15. Thus, we were unable to determine the type of 5-HT receptor involved in nighttime phase delays by quipazine or 5-HT. However, since the dose-response curves for quipazine during the day and night are virtually identical, we hypothesize that the nighttime 5-HT receptor is a 5-HT1-like receptor.