The fasciculus retroflexus controls the integrity of REM sleep by supporting the generation of hippocampal theta rhythm and rapid eye movements in rats.

The fasciculus retroflexus (FR) fiber bundle comprises the intense cholinergic projection from the medial division of the habenula nucleus (Hbn) of the epithalamus to the interpeduncular nucleus (IPN) of the limbic midbrain. Due to the widespread connections of the Hbn and IPN, it could be surmised that the FR is integrated in the processings of various subsystems that are known to be involved in the sleep-wake mechanisms; relevant sites include the limbic forebrain and midbrain areas and more caudal pontine structures. Consequently, the present study addressed the significance of the FR in the spontaneous sleep-wake stage-associated variations of the different activity patterns of frontal cortex and hippocampal electroencephalograms (EEGs), the electrooculogram, and body movements, in freely behaving rats that had been subjected to either bilateral electrolytic lesioning of the FR or control operations. The evolution of different state combinations was assessed by the combinatory analysis of different activity stages appearing on the 6-h records. As compared to the control-operated group, the FR lesioning substantially reduced the time spent in rapid eye movement (REM) sleep by 79%, moderately decreased the duration of the intermediate state of sleep by 29%, and quiet waking state by 44%, but had virtually no effects on the durations of different types of non-REM sleep (i.e., drowsiness that which involved quiet sleep or slow-wave sleep containing delta and spindle state components) or on the times of active waking behavior that corresponded to the body movements. Quantitative decomposition analyses revealed marked variations in the frontal cortex and hippocampal activity as well as REM during the course of the extracted sleep-wake stages described and there were also some group differences. Of those individual features that were used to determine different sleep-wake stages, the overall hippocampal theta time (41% decrease) and single REM frequency (71% reduction during the REM sleep) were most affected. In contrast, the various properties of desynchronization/synchronization patterns of frontal cortex EEGs were consistently hardly influenced by the FR lesioning. Therefore, the present data suggest the involvement of the FR in the REM sleep processes by establishing prominent associations with the limbic and REM control mechanisms that involve the hippocampus and plausibly pontine ocular activity networks.