Sleep Disorders — Functional Neuroanatomy
Sleep is probably the last complex and integrated behaviour of which the functions remain poorly understood. Nevertheless, since the last half of the 20th century, an impressive body of knowledge has been accumulated on sleep in mammals, especially cats and rodents. Much is now known on the mechanisms that maintain wakefulness, non-rapid eye movement (non-REM) sleep and rapid eye movement (REM) sleep, down to cellular and molecular levels.
In parallel, studies on patients and normal human subjects suggest that similar mechanisms generate sleep and sustain wakefulness in humans. However, we are still far from a comprehensive understanding of human sleep and its disorders at the fine-grained level of description reached in animal studies. Therefore, at present, it remains difficult to interpret the human pathology of sleep in terms of the underlying neuronal disturbances. Some exceptions, however, are mentioned in this chapter. For instance, recent breakthrough in molecular biology shed some light on the pathophysiology of narcolepsy in animals and, hopefully, in humans.
Sleep regulation does not involve only the generation of wakefulness and sleep periods. It also includes the interaction of sleep processes with the internal circadian rhythms and the external synchronizers, especially light. These aspects of sleep regulation are also important to consider because of the related human pathology.
This chapter provides an overview on the generation of sleep and wakefulness. It is not meant to review these mechanisms in the greatest detail but to put emphasis on the mechanisms that might be disturbed in sleep disorders. It is divided into two parts: (a) the regulation of circadian cycles and (b) the generation of wakefulness, non-REM sleep and REM sleep.
Sleep is not a homogeneous process. It is composed of two main sleep states: non-REM sleep and REM sleep. These two types of sleep differ in many aspects, such as their circadian distribution, their pattern of cellular activity and their physiological regulation. Non-REM sleep is usually recognized by the appearance of sleep spindles and K-complexes, the hallmarks of the light non-REM sleep (stage 2 sleep). The deepest stage of non-REM sleep, called slow-wave sleep (SWS), is defined by the presence of largeamplitude, low-frequency EEG waves. The power density of these slow waves is maximal at sleep onset and decreases exponentially during the night (Borbely, 1982).
Rapid eye movement sleep, also called paradoxical sleep (PS), is characterized by low amplitude, relatively fast rhythms on EEG recordings, ocular saccades and muscular atonia interspersed by muscular twitches. REM sleep typically occurs by periods of about 20 min, recurring every 90 min. In contrast to non-REM sleep, the duration of REM sleep episodes increases as the night progresses.
Sleep/waking cycles, like many other physiological and behavioural parameters, occur with circadian periodicity. In mammals, a large body of data proves that the suprachiasmatic nucleus (SCN) of the hypothalamus is the site which controls these circadian rhythms, especially sleep/wakefulness cycles.
First, in animals, lesions of the SCN abolish the rest/activity rhythmicity and alter sleep/waking patterns (Ibuka and Kawamura, 1975; Mouret et at., 1978). Second, circadian rhythms are reinstated by SCN transplants, the restored rhythms being derived from the donor SCN (Ralph and Lehman, 1991). Third, individual SCN cells maintain for long periods a near 24-h periodicity in their firing rates (Welsh et al., 1995). This observation suggests that the periodicity of the SCN activity does not emerge from local networks but reflects the genuine pacemaker properties of the SCN cells. This hypothesis was confirmed at the molecular level. The circadian rhythms are generated by autoregulatory feedback loops of transcription and translation of a set of clock genes and their gene products (for review, see King and Takahashi, 2000; WagerSmith and Kay, 2000). Manipulation of these clocks genes can lead to mutant individuals with intrinsic circadian rhythms different than the wild type (King and Takahashi, 2000; Vitaterna et al., 1994).
In addition, the clock of the SCN does not tick only on its own. It is entrained to the night/day cycle by light. Anatomically, SCN is in position both to receive light signal and synchronize several physiological rhythms. The SCN receives light signal from the retina, mainly through the retino-hypothalamic tract (Dai et al, 1998; Morin, 1994). It projects primarily to hypothalamic nuclei and, to a smaller extent, to limbic, thalamic and mesencephalic structures (van Esseveldt et al., 2000). The SCN can influence sleep/waking cycles via the secretion of melatonin, using a specific polysynaptic pathway (paraventricular nucleus, intermedio-lateral column of the spinal cord, superior cervical ganglion, and pineal gland) (see Moore, 1996).
Biological rhythms in humans seem to be generated by similar molecular and cellular mechanisms, which thus have important implications in sleep or circadian disorders. SCN is described in humans and, as in other mammals, it receives fibres from the retina and projects extensively to other hypothalamic nuclei (Dai et al., 1998). The disturbances of biological rhythms in demented patients have been attributed to structural modifications of the SCN (Swaab et al., 1985). As in other mammals, circadian rhythms in humans can be entrained by light (Lewy and Sack, 1996; Shanahan and Czeisler, 2000) and by melatonin (Cagnacci, 1997; Lewy and Sack, 1996). Blind people are known to suffer from sleep disorders due to free-running circadian rhythms (Nakagawa et al., 1992), with a circadian period rather longer than normal subjects (Sack et al., 2000). Melatonin may help these patients to entrain circadian rhythms and improve their condition (Sack et al., 2000). Finally, polymorphisms have been identified in several human clock genes.