Section of Psychopharmacology and Sleep Research
|
University of Zürich - Institute of Pharmacology and Toxicology Section of Psychopharmacology and Sleep Research |
Sleep interventions may have direct effects on slow-wave activity (SWA, i.e. power of the sleep EEG signal in the 0.75 - 4.5 Hz range) as well as indirect ones caused by changes in REM sleep (REMS) latency. The effects of changes in REMS latency on SWA were investigated by analyzing simulations with a mathematical model. Mean SWA in the first non-REMS episode shows an initial increase and a later decline as a function of REMS latency. In the second non- REMS episode, mean SWA decreases with increasing REMS latency. These results of the simulations were validated with experimental data. In the evaluation of the effects of sleep interventions on SWA the effects of the timing of REMS have to be accounted for. The analysis of SWA over a sufficiently long constant amount of time spent in non-REMS proves to be relatively independent of REMS latency, which allows conclusions about the effects of sleep interventions on SWA per se.
J. Sleep Res. 4 (1995): 23-29.
Sleep inertia refers to the period of reduced vigilance following upon awakening from sleep. To investigate the time course of sleep inertia, self- ratings of alertness and reaction time in a memory task were repeatedly assessed after nighttime and daytime sleep episodes in healthy young men. Alertness gradually increased and reaction time gradually decreased within the first hour after awakening. Their time course could be described by exponential functions with time constants of 0.45 h and 0.3 h, respectively. The data demonstrate that sleep inertia is a robust, quantifiable process that can be incorporated in models of sleep and vigilance. [References: 35]
Arch. Ital. Biol. 134 (1995): 109-119.
Djungarian hamsters well adapted to a short photoperiod were subjected to 4 h of total sleep deprivation (SD) by gentle handling. Tissue concentrations of monoamines and of their metabolites were measured from several brain areas using HPLC with electrochemical detection. The 5-hydroxyindoleacetic acid/5- hydroxytryptamine (5-HIAA/5-HT) ratio was significantly increased after SD in the hippocampus, hypothalamus and brain stem, indicating increased serotonin (5- HT) turnover in those areas, while no changes were found in the frontal cortex and olfactory bulb. Dopamine and 3,4-dihydroxyphenylacetic acid (DOPAC) concentrations were elevated in the hypothalamus, while the noradrenaline concentrations did not change in any of the measured areas. We conclude that a short SD, which has been shown to elevate EEG slow-wave activity during recovery sleep, specifically increases 5-HT turnover in the brain. [References: 24]
Neurosci. Lett. 198 (1995): 21-24.
We have investigated the effects of changes in brain temperature on the electroencephalogram (EEG) during entrance into daily torpor, a natural hypothermic state, in the Djungarian hamster. A systematic shift of single EEG frequencies was found as cortical temperature decreased. The relation between EEG frequency and cortical temperature was very similar to the temperature dependence of the Na(+)-K(+)-pump, suggesting that the pump is the rate-limiting step in determining EEG frequency.
Brain Res. 670 (1995): 153-156.
Slow waves and sleep spindles are prominent features of the EEG in non- REM sleep and some of the neurophysiological mechanisms underlying their genesis have been elucidated. In humans, slow-wave activity in non-REM sleep increases and EEG activity in the frequency range of sleep spindles decreases when wakefulness prior to sleep is varied from 2 to 40 h. The opposite changes are observed in the course of sleep, even when sleep is scheduled out of phase with the circadian rhythm of sleep propensity. Within non-REM sleep episodes the association between slow waves and sleep spindles is bi-phasic: both activities are correlated positively at the beginning and end of non-REM sleep episodes whereas in the middle part of non-REM sleep episodes high values of slow-wave activity coincide with low levels of spindle activity. An extension of wakefulness enhances the rise rate of slow-wave and spindle activity at the onset of sleep. Since macroscopic slow waves and sleep spindles both are dependent on hyperpolarization and synchronization of neurons in thalamo- cortical and cortical circuits, the sleep deprivation induced changes in these EEG activities may be related to reduced activating input to thalamo-cortical and cortical neurons, local facilitation of their hyperpolarization or facilitation of their synchronization. The precise regulation of slow-wave and spindle activity as a function of the duration and intensity of prior sleep and wakefulness demonstrates that these BEG oscillations are accurate indicators of non-REM-sleep homeostasis and suggests that they are fundamental to the sleeping brain. [References: 63]
Behav. Brain Res. 69 (1995): 109-116.
The effect of melatonin (5 mg, p.o.) on electroencephalographic (EEG) activity during sleep was investigated in eight men in a placebo-controlled cross-over design. Melatonin was administered immediately prior to a 4-h daytime sleep episode (13-17 h) after a partial sleep deprivation. The non-REM sleep stages and REM sleep duration were not significantly affected. Melatonin enhanced EEG power density in non-REM sleep in the 13.75-14.0 Hz bin (i.e., within the frequency range of sleep spindles), and reduced activity in the 15.25-16.5 Hz band. In the first 2 h spectral values within the 2.25-5.0 Hz range were reduced. These changes in the EEG are to some extent similar to those induced by benzodiazepine hypnotics and to the contribution of the endogenous circadian pacemaker to the spectral composition of the sleep EEG when sleep occurs at night. [References: 33]
Neurosci. Lett. 201 (1995): 13-16.
To assess the influence of the photoperiod on sleep regulation, laboratory rats were adapted to a long photoperiod (LPP; 16:8-h light-dark cycle, LD 16:8) or a short photoperiod (SPP; LD 8:16). The electroencephalogram (EEG) and cortical temperature (T-CRT) were continuously recorded for a baseline day, a 24-h sleep deprivation (SD) period, and a recovery day. Data obtained previously for LD 12:12 served for comparison. Whereas the photoperiod exerted a prominent effect on the 24-h sleep pattern, the 24-h baseline level of sleep and the response to SD were little affected. Recovery from SD was characterized by a marked rise in rapid eye movement sleep, a moderate rise in non-rapid eye movement sleep, and an initial enhancement of EEG slow-wave activity followed by a decrease below baseline. The amplitude and phase of the "unmasked" 24-h component of T-CRT did not differ between LPP and SPP. Computer simulations demonstrated that the changes of T-CRT and EEG slow-wave activity can be largely accounted for by the sequence of the vigilance states. We conclude that the photoperiod does not affect the basic processes underlying sleep regulation. [References: 33]
Am. J. Physiol. 269 (1995): R691-R701.
To examine the relationship between sleep and brain temperature in the rat, the vigilance states, spectral power density of the electroencephalogram (EEG), hypothalamic temperature (T(hy)), and cortical temperature (Tcr) were recorded for 3 days. A 1-day rise of ambient temperature from 23 to 30 degrees C did not affect the percentage of waking, non-rapid eye movement sleep (NREMS), and rapid eye movement sleep (REMS), but increased EEG slow-wave activity in NREMS in the 12-h dark period. T(hy) was invariably higher than Tcr, but at 30 degrees C the difference diminished because of a rise in Tcr. In contrast to Tcr, T(hy) was only slightly increased at 30 degrees C and only during sleep and in the dark period. Although the temperatures changed largely in parallel at vigilance state transitions, Tcr rose more rapidly than T(hy) at NREMS-REMS transitions and more slowly at NREMS-waking transitions. T(hy) declined more rapidly than Tcr at waking-NREMS transitions and more slowly at REMS-NREMS transitions. The results are consistent with a central role of the hypothalamus in the activation and deactivation of the waking state.
Am. J. Physiol. 268 (1995): R1365-R1373.
In view of the hypothesis that adenosine is involved in sleep regulation, the effects of the adenosine antagonist caffeine on sleep and sleep EEG were investigated in eight young males. Compared to the placebo condition, caffeine (100 mg) administered at bedtime prolonged sleep latency and reduced sleep efficiency and stage 4 of non-rapid eye movement sleep (NREMS). Electroencephalographic slow-wave activity (SWA, spectral power density in the 1.75-4.5-Hz band) was reduced, whereas power density in the spindle frequency range was slightly enhanced. The suppression of SWA was limited to the first NREMS episode. Caffeine reduced the power density mainly in the lowest delta band, in contrast to the changes during physiological sleep that encompass both the delta and theta bands. Caffeine levels in saliva, assessed in a separate experiment, decreased from 7.5 mumol/l in the first hour of sleep to 3.5 mumol/l in the seventh hour. In the night following caffeine administration, stage 4 sleep had reverted to the baseline level, but sleep latency was still increased, and stage 2 sleep, as well as SWA in the first NREMS episode, were reduced. The data show that even a low dose of caffeine affects the sleep EEG. However, the effects of caffeine did not completely mimic the spectral changes observed during physiological sleep.
Neuropsychopharmacology 12 (1995): 229-238.
The 24 h time course of intracranial temperature, recorded subdurally at the parahippocampal gyrus in six patients (19 24 h periods), exhibited a prominent 24 h rhythm with its crest located at 20-21 h. The declining trend of intracranial temperature between lights off and sleep onset persisted in the first nonREM sleep episode (studied in two patients, seven sleep episodes). The correlation between EEG slow-wave activity (SWA) in nonREM sleep and the change in temperature explained < 25% of the variance. Although the change in temperature tended to be more positive in REM sleep episodes than in nonREM sleep episodes, no significant increase was observed in REM sleep. The data indicate that intracranial temperature exhibits a marked 24 h rhythm, the time course of which is only slightly affected by nonREM/REM sleep and EEG synchronization.
Neuroreport 6 (1995): 913-917.
Adenosine has been implicated in the physiological regulation of sleep propensity. The adenosine-receptor-antagonist, caffeine (100 mg), administered immediately prior to a nocturnal sleep episode, has previously been shown to lower sleep propensity as indexed by a reduced sleep efficiency, a reduced EEG power density in low delta frequencies and enhanced power density in the frequency range of sleep spindles. To further investigate the role of adenosine in sleep regulation we administered 200 mg of caffeine at 07.10 h and analyzed the sleep stages and EEG power spectra during the subsequent night in nine healthy men. Caffeine levels in saliva decreased from a maximum of 17 mumol/l one hour after intake, to 3 mumol/l immediately prior to the sleep episode starting at 23.00 h. Compared to placebo, sleep efficiency and total sleep time were significantly reduced. EEG power density in nonREM sleep was suppressed in the 0.25-0.5 Hz band and enhanced in the frequency range of sleep spindles (11.25-12.0 Hz and 13.25-14.0 Hz). In REM sleep EEG power density was suppressed in the frequency range of 0.75-4.5 and 5.25-6.0 Hz. The data indicate that a saliva level of caffeine as low as 3 mumol/l directly affects sleep propensity or, alternatively, that the presence of caffeine in the central nervous system during the waking episode reduces the progressive increase of sleep propensity associated with wakefulness.
Brain Res. 675 (1995): 67-74.
The mRNA level of the 17-kDa protein neurogranin (NG), a postsynaptic substrate of the protein kinase C, has previously been found to be decreased in rat forebrain after 24-h sleep deprivation (SD). To investigate the functional significance of this finding in various forebrain regions, the effect of 24-h SD on the mRNA level and the protein level of NG was determined in the cerebral cortex, hippocampus, and the total of the remaining subcortical forebrain plus midbrain areas (SFMA) of rats. In these areas, high levels of both NG mRNA and NG protein were detected by in situ hybridization and immunohistochemistry, respectively. NG protein was recognized in brain tissue by newly developed polyclonal antibodies. As determined by RNase protection assays, the level of NG mRNA was decreased in SFMA by 34 +/- 7% (P < 0.05) after 24-h SD, and was not significantly affected in the cerebral cortex and hippocampus. In contrast, on Western blots, the protein concentration of NG was reduced in the cerebral cortex by 37 +/- 7% (P < 0.05) whereas no significant changes were present in other brain areas tested. The results indicate that the mRNA and protein levels of NG are differentially modulated in rat brain by the prolongation of the waking period. [References: 40]
Brain Research 685 (1995): 143-153.
An avenue to investigate the functions of sleep is the comparison of sleep in different species, particularly in closely related ones and in species with extreme specializations. The features which are usually investigated are the occurrence of both sleep stages non-REM sleep and REM sleep, their amount per 24 h, the duration of the non-REM-REM sleep cycle and the daily distribution of sleep relative to the light-dark cycle of the environment. Recently also sleep homeostasis has been included, because it is now well established that mammalian species can compensate for sleep loss both by an increase in sleep duration as well as by intensifying non-REM sleep. The occurrence of EEG slow- wave activity has served as a measure for sleep intensity. The capacity to sleep more intensely enables animals to react more flexibly to sleep loss. The comparison of mammalian species has revealed striking similarities in the way sleep is regulated which indicates common underlying mechanisms. [References: 55]
Behav. Brain Res. 69 (1995): 35-41.
The aim of the study was to examine whether the typical changes of the EEG in the course of a sleep episode can be recorded by skin electrodes placed at the outer canthi of the eyes. In sleep recording from young, healthy subjects, the signals from the electro-oculogram (EOG, E1-A2) derivation and a central scalp EEG derivation (C3-A2) were compared. Sleep stage scores obtained separately from each of the two signals yielded highly corresponding values with the exception of stages 2 and 4, for which there were discrepancies. Both signals were subjected to spectral analysis to compare the spectra and their evolution during sleep. An automated detection routine served to identify and eliminate epochs contaminated by eye movement potentials. The typical time course of EEG slow-wave activity (SWA; power density in the 0.75-4.5 Hz range) could be derived from both signals with only minor differences between the data sets. However, compared to the EEG spectra, power density of the EOG spectra was attenuated in frequencies higher than 2 Hz and the typical changes in the spindle frequency range were not evident. The results show that the major sleep parameters as well as the dynamics of SWA can be reliably determined from signals recorded from periorbital skin electrodes.
Electroencephalogr. Clin. Neurophysiol. 94 (1995): 406-413.