Summary

Since the discovery of rapid eye movement (REM) sleep (also known as paradoxical sleep, PS), it has been accepted that sleep is an active process. Paradoxical sleep is characterized by electroencephalogram (EEG) rhythmic activity resembling that of waking with a disappearance of muscle tone and the occurrence of REMs in contrast to slow-wave sleep (SWS, also known as non-REM sleep) identified by the presence of delta waves. Here, we review the most recent data indicating that glutamatergic neurons play a key role in the genesis of PS. We propose an updated integrated model of the mechanisms responsible for PS integrating these neurons. We hypothesize that the entrance from SWS to PS is due to the activation of PS-active glutamatergic neurons localized in the pontine sublaterodorsal tegmental nucleus (SLD). We further propose that these neurons are tonically excited across all the sleep–waking cycle by glutamatergic neurons localized in the lateral periaqueductal gray. We finally hypothesize that the onset of activity of the SLD glutamatergic neurons is due to the removal of a GABAergic input from neurons localized in the ventrolateral periaqueductal gray and the adjacent deep mesencephalic reticular nucleus.

Paradoxical sleep: the early years

In 1959, Michel Jouvet and François Michel discovered in cats a phase of sleep characterized by a complete disappearance of the muscle tone, and paradoxically associated with a cortical activation and rapid eye movements (REM) (Jouvet & Michel, 1959). In view of its singularity, they proposed to call this state paradoxical sleep (PS). It corresponded to REM sleep, the state described in 1953 by Aserinsky and Kleitman and that correlates with dream activity in humans (Aserinsky & Kleitman, 1953; Dement & Kleitman, 1957). In view of the occurrence of muscle atonia, Jouvet proposed that PS was a distinct sleep state and a true vigilance state independent of slow wave sleep (SWS) and waking (W). Over the 40 years following its discovery, Jouvet and co-workers pursued the study of PS. Supporting the theory of the duality of sleep, they showed that PS, in contrast to SWS, was present in mammals and birds but absent from amphibians and reptiles. Jouvet also demonstrated that PS onset and maintenance depended upon structures different from those regulating SWS and W. He first showed that PS persists after decortication, after cerebellar ablation or transections of the brainstem rostral to the pons. In contrast, transections at the posterior limit of the pons suppressed PS (Jouvet, 1962a).