Sleep is a global state, the control mechanisms of which are manifested at every level of biological organization, from genes and intracellular mechanisms to networks of cell populations, and to all central neuronal systems at the organismic level, including those that control movement, arousal, autonomic functions, behavior and cognition.Recent genetic findings indicate that the molecular mechanisms that control circadian rhythms, which set the stage for sleep and are inseparable from sleep in a deep biological sense, are highly conserved phylogenetically. Molecular and behavioral conservation indicates that sleep conferred a selective advantage on ancestral mammals that might persist in modern populations.Prolonged sleep loss impairs temperature control, dietary metabolism and immune function, and leads ultimately to death.In the mammalian nervous system, genetic instructions are expressed at the progressively higher levels of gene transcription, protein synthesis and intracellular events, individual neuronal and neuronal-network dynamics, and ultimately behavior, of which cognition is a specific covert form. In this review, the genetic mechanisms, cellular neurophysiology and subcortical neuronal-population networks are discussed that are involved in sleep.
(Albrecht, 2002)The circadian pacemaker Starting with the first demonstration of a circadian gene in the fruitfully, genetic approaches have begun to illuminate the intranuclear and cytoplasmic events that are associated with circadian rhythms and sleep. Most notable has been the elucidation of the genetic control of the mammalian circadian pacemaker, which can now explain the near-perfect 24-h rhythmicity of the human circadian clock.Mammalian circadian rhythms are maintained intracellularly by interlocking positive- and negative-feedback control of the transcription (and subsequent translation to protein) of three period genes (Per1–3), two cryptochrome genes (Cry), and the Clock and Bmal (brain and muscle ARNT-like 1) genes. The products of Clock and Bmal exist as a heterodimer that is a key component of a transcription factor (abbreviated Clock: Bmal) that promotes the transcription of per and Cry genes by binding to their regulatory DNA sequences (E-box elements). The Per and Cry messenger RNAs are translocated to the cytoplasm for translation to proteins that form complexes that then re-enter the nucleus to exert feedback control on the Clock:Bmal transcription factor. Products of several other genes modulate this intracellular mechanism. For example, the product of the tau (Csnk1e) gene, casein kinase 1?, phosphorylates per proteins, which affects their translocation between the cytoplasm and the nucleus.Briefly, molecular feedback control of the circadian clock in SCN neurons operates as follows.
At the beginning of the organism’s SUBJECTIVE DAY, CIRCADIAN TIME (CT), the transcription and translation of Per and Cry are accelerated by Clock: Bmal heterodimers that have accumulated over the previous subjective night (CT 12–24). The levels of the Per and Cry complexes peak at the beginning of the organism’s subjective night (CT 12).Protein complexes that contain the products of the Cry gene exert negative feedback on the Clock:Bmal promoter, thereby slowing the transcription of Per and Cry. At the same time, a protein complex that contains Per exerts positive feedback by promoting the transcription of Bmal.The combined action of positive- and negative feedback loops creates a suite of molecular signals that reliably recur at precise times over 24-h cycles. These molecular signals can be read by cytoplasmic mechanisms in SCN cells and translated into reliably recurring cellular events, such as changes in membrane potential.
Such signals, in turn, can be transmitted to connecting neurons and, ultimately, to those neural structures that control physiological processes with a circadian rhythmicity. Many of the details of the genetic mechanisms of mammalian circadian rhythms have been elucidated by studies of mutations in these genes. For example, a mouse mutant of the Clock gene shows lengthening of the circadian period. Cloning of these mammalian genes, beginning with Clock, has allowed the precise molecular analysis of normal and mutant circadian genes.
A recent study in humans has reported a heritable, familial trait for advanced sleep-phase syndrome with autosomal-dominant transmission. This finding might constitute the first step in identifying molecular components of the human circadian system that are analogous to those described in animals.The induction of Per might be the first step in resetting the SCN clock. Like membrane receptors and intracellular pathways, such intranuclear events also show specific sensitivity to circadian phase.
For example, CREB phosphorylation occurs only during the night.SCN neurons communicate circadian time to other brain structures primarily by action potentials that are mediated by sodium channels, and most SCN cells contain the inhibitory transmitter GABA (?-amino butyric acid).However, the circadian clock continues to keep accurate time even when SCN cell firing is blocked by TETRODOTOXIN. SCN firing peaks at the middle of the circadian day, and circadian variations in SCN action potentials are believed to be produced by similar variations in SCN cell-membrane properties.Surface rhythms, such as variation in membrane potential or ion channel activity, are likely to be controlled by proteins such as prepropressophysin (a precursor of arginine vasopressin, AVP) — products of ‘clock-controlled genes’ that are under the transcriptional control of core clock genes, such as Clock and Bmal.SCN neurons probably synchronize primarily by using GABA, although diffusible substances and cell-surface constituents might also have non-synaptic roles. Peripheral circadian oscillators that are molecularly similar to those in the SCN also exist throughout the body. However, current evidence indicates that the maintenance of these oscillators depends on periodic input from the SCN.
Narcolepsy. The important role of genetics in sleep research is also exemplified by the discovery of the genetic basis of narcolepsy. These insights have led to the discovery of an excitatory wake-promoting neuromodulatory system that originates in the orexin (or hypocretin)-producing cells of the lateral hypothalamus.
Gene expression during sleep and wakefulnessImmediate early genes (IEGs), such as c-fos, are reliably transcribed and translated to protein products shortly after a neuron becomes active. Their products are probably involved in the subsequent transcription of other genes. Regional expression of these genes is highly state dependent, and most brain cells express c-fos and other IEGs at higher levels during waking.Other genes that are activated selectively during sleep or waking might encode proteins that, unlike IEGs, are involved specifically in state-dependent processes. Extensive screening of the rat genome has been undertaken recently to identify genes that are upregulated selectively during sleep or waking. Using cortical cells of sleeping, waking and sleep-deprived rats, ~30–75% of the rat genome has been screened.
State-dependent changes in regulation were found in less than 1% of the screened genome, and most of these genes were upregulated selectively during normal or sleep-deprived wakefulness.The functions of the few rat genes that are upregulated selectively in sleep are unknown. Genes that are upregulated selectively during short (3 h) periods of wakefulness include IEGs and related transcription factors, as well as components of the mitochondrial genome, including the gene for subunit I of cytochrome coxidase, an enzyme that is involved in oxidative metabolism. Such wake-related activation of the mitochondrial genome might facilitate a rapid response to the metabolic demands of initial, brief or unpredictable wakefulness, as the expression of these genes returns to baseline levels after sustained wakefulness.Genes that are upregulated during sustained waking include some that are involved in glucose metabolism and responses to physiological stress. Most interesting to us are genes that encode elements of synaptic neurotransmission, such as presynaptic transporters and postsynaptic receptors.
Upregulation of these genes could reflect a wake-related increase in demand that is associated with synaptic efficacy and neuroplastic processes. For example, the expression of aryl sulphotransferase, which is involved in the breakdown of the wake-related catecholamine neurotransmitters, is upregulated during extended sleep deprivation.Notwithstanding this controversy, Chou, et al (2002) present compelling arguments that changes in the activity of the PKA second-messenger system and its third messenger, CREB, are associated with both state dependent neuromodulatory changes in the hippocampus and performance of hippocampus-dependent learning tasks. Consolidation of hippocampus-based learning in rats is sensitive to disruption at specific times after training by both REM sleep deprivation and the intraventricular injection of inhibitors of PKA or the subcutaneous injection of inhibitors of protein synthesis. These manipulations could disrupt a PKA ?CREB ?gene transcription ?protein synthesis pathway that is necessary for the consolidation of learning and is specifically facilitated by REM sleep.
The hypothetical mechanism of this facilitation isREM-related enhanced cholinergic and diminished serotonergic modulation of adenylyl cyclase activity (linked to specific acetylcholine (ACh) and serotonin membrane receptors), which determines the activity of the PKA signaling pathway. Notably, rats that are exposed to rich sensorimotor experiences during waking have elevated levels of the plasticity-associated IEG zif-268 (also known as early growth response during subsequent sleep. Cellular Neurophysiology of REM and NREMThe original reciprocal-interaction model proposed that aminergic and cholinergic neurons of the mesopontine junction interact in a manner that results in the ULTRADIAN alternation of mammalian REM and NREM sleep.
In this model, REM-on cells of the pontine reticular formation are cholinoceptively excited postsynaptically and/or cholinergically excitatory at their synaptic endings. Pontine REM-off cells are noradrenergically or serotonergically inhibitory. During waking, the pontine aminergic system is TONICALLY activated and inhibits the pontine cholinergic system.Recent evidence supports this model, but intermediate synaptic steps might intervene in the initiation and augmentation of REM at the level of both REM-on mesopontine neurons and REM-off pontine aminergic nuclei. Such synaptic details can be integrated with the reciprocal-interaction model without altering the basic effects of aminergic and cholinergic influences on the REM sleep cycle.Networks that generate the signs of REM sleep. A distributed neuronal network generates the characteristic signs of REM sleep. The initiation of REM sleep- related activity in these networks results from changes in the activity of ‘executive’ neuronal populations in the mesopontine junction, as identified in the reciprocal-interaction model.
The net result of REM-related activity in executive neuronal populations is the strong tonic and phasic activation of brainstem reticular and sensorimotor relay neurons in REM sleep. The characteristic signs of REM sleep are postulated to be further mediated as follows:• EEG activation results from a net tonic increase in reticular, thalamocortical and cortical neuronal firing rates.• Phasic potentials that are recorded sequentially in the pons, thalamic lateral geniculate body and occipital cortex of the cat, termed ponto-geniculo-occipital (PGO) waves, are the result of tonic disinhibition and phasic excitation of burst cells in the lateral pontomesencephalic tegmentum.• REMs are the result of phasic firing by reticular and vestibular cells; the latter directly excite oculomotor neurons.• The change in the hippocampal EEG from irregular rhythms to a regular THETA RHYTHM is influenced by the brainstem and mediated by the medial septal nucleus of the basal forebrain.• Muscular atonia results from tonic postsynaptic inhibition of spinal anterior horn cells by the pontomedullary reticular formation.
NO has been implicated widely in sleep-cycle modulation, and functions primarily as an intercellular messenger that can enhance capillary vasodilation and the synaptic release of neurotransmitters such as ACh. It is produced by cholinergic mesopontine neurons, and might help to maintain the cholinergically mediated REM sleep state in the pons and thalamus.The level of extracellular NO is elevated when the activity of mesopontine cholinergic neurons increases, and NO modulates the release of ACh in the basal forebrain. (Rye, & Jankovic, 2002)The vascular effects of NO might interact with statedependent alterations in cholinergic excitability and ascending cholinergic activation,which, in turn, could produce the REM-related changes in regional blood flow that are seen in neuroimaging studies.Finally, neuropeptides such as vasoactive intestinal polypeptide (VIP), as well as numerous hormones, are increasingly thought to regulate REM–NREM cycles. Research on the REM–NREM cycle is now beginning to extend from the neurotransmitters and their receptors to the roles of intracellular second messengers and the molecular biology of gene transcription.The anterior sleep-promoting and posterior arousal promoting regions of the hypothalamus are thought to be mutually inhibitory. The wake-promoting neuromodulators noradrenaline, serotonin and AChinhibit VLPO neurons.
Although histamine has not been shown to inhibit VLPO neurons, histaminergic neurons are co-localized in the TMN with GABA neurons, which might reciprocally inhibit the GABA containing VLPO neurons that project to the histaminergic TMN cells. In support of these ideas, serotonergic dorsal raphe, noradrenergic locus coeruleus and histaminergic TMN neurons project to the VLPO. Hattar et al (2002) propose that this arrangement forms the dynamic basis for a pattern of bi-stability analogous to a ‘flip–flop’ electrical circuit. Either sleep or the waking state is self-reinforcing when its component neurons are sufficiently active; transitional states arise when either wake or sleep neuronal activity wanes, but they are transient because the system tends to revert to one of the two stable configurations. The orexin system of the lateral hypothalamus. The orexin system — an important system for the control of behavioral state — was identified by investigators seeking the genetic basis of narcolepsy. In the rat, orexin induces a dose-dependent increase in wakefulness when it is injected intracerebroventricularly or perused by microdialysis into the basal forebrain. The densest projection of orexinergic cells in the rat is to the locus coeruleus, the noradrenergic output of which favors the cortical arousal of waking, but opposes REM associated arousal.
In humans, orexinergic neurons in the perifornical, lateral and medial hypothalamus project heavily to the locus coeruleus. Interestingly, human narcoleptics show a large reduction in the number of orexinergic neurons in the lateral hypothalamus and a deficiency of orexin in the cerebrospinal fluid.On the basis of the bistable hypothalamic sleep–wake switch model, Lu, Xu, & Saper, (2002) have proposed a specific functional role for orexin in the normal sleep–wake cycle that might explain the dysregulated state transitions that occur in its absence in narcolepsy .They propose that the orexinergic drive on key nuclei in the wake-promoting, aminergic half of the bistable sleep–wake switch stabilizes the wake state and prevents untimely transitions from waking to sleep. In narcolepsy, orexinergic deficiencies make abnormal wake–sleep switches more likely.
Orexin might indirectly inhibit sleep-promoting neurons of the VLPO by exciting co-localized inhibitory wake-on neurons in the preoptic hypothalamus. Clearly, this important modulator of primary sleep–wake and REM–NREM mechanisms will continue to be a focus of intense study. (Hannibal et al, 2002)