Genetics of Sleep and Sleep Disorders
The relationship of genes to behaviour has always been a major topic of research and interest in the neurosciences. In the study of sleep, the traditional way to determine whether distinct phenotypes have a familial component was to study family pedigrees or to investigate sleep in monozygotic (MZ) and dizygotic (DZ) twins, a classical method in human and psychological research. In animals, the inheritance of sleep traits was previously studied by Valatx et al. (1972) in cross-breeding mice strains. The developments of molecular genetics and gene technology favoured a renewed interest in the behavioural neurobiology of sleep, leading to an increasing number of papers on the genetics of sleep and the molecular aspects of sleep disorders.
Current scientific investigations address various phenotypic dimensions, including the genetic components of rapid-eye-movement (REM) sleep and delta sleep, and the circadian components of human sleep.
In the present paper we will review some of the classical genetic studies of sleep and summarize important recent results.
The importance of the investigation of the genetic aspects of sleep was demonstrated in some early studies that focused mainly on the investigation of twin pairs in man. In 1937, Geyer (quoted in Vogel, 1958) reported that concordance for a number of sleep characteristics was much better in MZ than in DZ twins. Other pioneering efforts were made by Vogel (1958), who was the first to study the sleep EEG in MZ twins, and by Zung and Wilson (1967), who performed computations of the sleep EEG of three pairs of MZ and three pairs of DZ twins. They reported that the all-night-sleep EEG and REM patterns of MZ twins were similar, whereas those of DZ twins were dissimilar and variable.
It should be noted, however, that some further studies relied mainly on questionnaire investigations of sleep habits (duration of sleep, subjective reports of sleep timing and quality) in MZ and DZ twins (Partinen et al, 1983; Heath et al, 1990). These studies were conducted mainly in adult twins, and, in general, correlations were higher for MZ than for DZ twins, with a significant synchronizing effect of cohabitational status on sleep length (Partinen et al, 1983). Heath et al (1990) found that genetic influences accounted for at least 33% of the variance in sleep quality and sleep disturbance, and 40% of the variance in sleep pattern. A substantial genetic effect was also reported for some parasomnias such as sleepwalking and nocturnal enuresis (Kales et al., 1980).
The hereditary components of sleep were also suggested by some polygraphic twin studies of newborns (Gould et al., 1978), of a limited number of adolescent twin pairs, and of single nights of sidentical and fraternal twins. In the study of Webb and Campbell (1983), measures of awakening and REM sleep amounts were found to be significantly correlated in identical twins. In a study of seven twin pairs, Hori (1986) suggested that sleep spindles were determined by a genetic trait.
The inheritance of properties of the wake EEG has also been a focus of interest in recent years, since the pioneering observations of Vogel (1958). Stassen et al (1987) reviewed 22 twin studies and noted that MZ co-twins were generally more alike than DZ co-twins, indicating the presence of genetic influences on the spontaneous EEG. In a more recent study, Van Beijsterveldt et al. (1996) measured four frequency bands (delta, theta, alpha, and beta) and reported that the average heritability of brain functioning, as indexed by rhythmic brain electrical activity, is one of the most heritable characteristics in humans.
Finally, a large number of animal studies also indicate unequivocally that certain aspects of sleep might be genetically determined.
These studies have mainly suggested a genetic determination of sleep length in rodents (Valatx et al., 1972; Friedman, 1974; Franken et al, 1999) as well as a significant genetic effect on REM sleep in mice (Valatx et al, 1980; Valatx, 1984).
Taken together, most studies that addressed the issue of genetic aspects of sleep in normal men have relied on the twin method, which allows the determination of heritability estimates and is frequently used to partition variance of phenotypic quantitative traits into environmental and genetic components.
The twin design has also been frequently used to evaluate the heritability of other primary sleep disorders such as sleepiness, nightmares and sleepwalking. This will be reviewed later in the chapter. We will first present the results of the Brussels twin study, which was conducted between 1986 and 1990 (see Table XXIV-9.1) (Linkowski et al., 1989; 1991).
In this study of two samples of 26 pairs of male twins, we have found that stages 2, 3, 3 + 4 and 4 were significantly determined by genetic components. This was also the case for REM density. No genetic effect was found for total sleep period, period of sleep, total sleep time, sleep onset latency and REM latency. Waking was found to be significantly determined by genetic factors in one of the two samples.
Therefore, our results substantiate previous reports suggesting that some components of human sleep might be genetically determined (Partinen et al, 1983; Webb and Campbell, 1983). In our study, stages 2, 4 and delta sleep (stages 3 + 4) unequivocally show a strong genetic component: the same is true, but is less pronounced for REM density measures. It is of interest that the stages with the strongest genetic component show the best relative stability from night to night (Merica and Gaillard, 1985). This is also supported by previous reports of substantial within-subject, night-to-night correlation of stage 4 (Webb and Agnew, 1968) and a greater inter- than intraindividual variation of stage 4 (Merica and Gaillard, 1985).