During REM sleep, the brain exhibits electrical activity typical of a highly aroused state (desynchronized cortical EEG, hippocampal theta rhythm, rapid eye movements, variable breathing and arterial blood pressure) while motor activity is suppressed and one is unaware of the external environment. Wake-active neuronal groups (serotonergic, noradrenergic and histaminergic) are silent, while some cholinergic neurons have high levels of activity. Dreams occur during this stage of sleep. Neuronal networks localized in the pons and medulla are sufficient for the generation of REM sleep but suprabrainstem centers (e.g., hypothalamus, basal forebrain) powerfully modulate its occurrence and expression. Under the normal sleep-wake conditions, the activity of wake-related neurons is greatly reduced during slow-wave sleep; this facilitates the subsequent occurrence of REM sleep. Activation of a distinct group of hypothalamic wake-active cells containing the excitatory peptides orexins (also called hypocretins) suppresses the ability of the brainstem network to generate REM sleep. The evidence for this is based on the observation that humans, dogs and mice with disrupted synthesis of orexins or their receptors have symptoms of narcolepsy/cataplexy, a disorder characterized by an "out of context" (e.g., during wakefulness) activation of the key components of the REM sleep network. When forebrain influences are suppressed or eliminated, the brainstem can relatively freely generate multiple REM sleep-like episodes either spontaneously or in response to certain external triggers.
Many electrophysiological signature events of REM sleep can be initiated in experimental animals pharmacologically by local microinjections of a cholinergic agonist, carbachol into a discrete region of the pontine reticular formation. Such a pharmacologically elicited state has a stereotyped pattern similar to natural REM sleep and can be elicited in unanesthetized, decerebrate animals, in anesthetized animals (Kubin, Arch. Ital. Biol. 139:147-168, 2001; full text: HTML/PDF) and even in the highly reduced preparation of isolated and artificially perfused brainstem-spinal cord (Brandes et al., Exp. Physiol. 96:548-555; 2011; see: [http://dx.doi.org/10.1113/expphysiol.2010.056242). In decerebrate cats, pontine injections of carbachol produce postural atonia, depression of upper airway muscle tone, and eye movements. In urethane-anesthetized rats, pontine carbachol can repeatedly trigger short (2-4 min) episodes of cortical activation, hippocampal theta-like rhythm and suppression of activity in upper airway (hypoglossal) motoneurons. REM sleep-like episodes elicited in the urethane-anesthetized rat model are associated with silencing of noradrenergic cells of the A5 group, a group important for cardiorespiratory regulation (Fenik et al., J. Appl. Physiol., 93: 1448-1456, 2002), as well as the noradrenergic cells of the locus coeruleus.
Medullary serotonin-containing cells that send axons to the hypoglossal motor nucleus are also silenced during the period of postural atonia and depression of XII nerve activity associated with postural atonia triggered by pontine carbachol (Woch et al., J. Physiol. (Lond.) 490: 745-758, 1996). On the other hand, ventrolateral medullary adrenergic neurons are activated during REM sleep-like episodes elicited by pontine carbachol in urethane-anesthetized rats (Stettner et al., PLoS ONE 8(4): e62410, 2013; full text: PDF). This may explain the characteristic for REM sleep elevation and variability of arterial blood pressure.
Hence, there is extensive electrophysiological evidence that many of the neural phenomena that occur in reduced carbachol models of REM sleep adequately represent the corresponding neural events characteristic of the natural REM sleep. The ability to produce REM sleep-like phenomena by microinjections of carbachol into a discrete region of the dorsomedial pontine tegmentum offers an attractive tool with which to study cellular behaviors associated with REM sleep under rigidly controlled experimental conditions.