Abstract
Sleep regulation is realized by homeostatic and circadian processes. The homeostatic component is reflected by slow-wave activity (SWA) in the NREM sleep EEG, while high frequency activity (>10 Hz, HFA) is thought to be influenced by the circadian factors. We investigated whether daily changes in sleep and HFA are independent of changes in SWA. Rats (n=4) were implanted with EEG and EMG electrodes and kept in 12h:12h light/dark cycles for the recovery period. Subsequently the animals adapted to constant dark conditions (DD) for at least 1 week. A baseline (BL) day was recorded followed by “short-day protocol” - 2h sleep deprivation followed by 2 h rest for 2 days. The vigilance states were determined and EEG spectral analysis between 0.1-25.0 Hz was performed. The amount of sleep over 24 h of the “short-day protocol” was less (43.3±1.7% SE) compared to baseline (51.5 ±1.2%, p<0.005, t-test). Circadian changes in vigilance states was reduced to 25 % of baseline (p<0.05, t-test). SWA (1.1-4.0 Hz) in NREM sleep did not show a significant circadian modulation (p>0.5, ANOVA 4-h intervals) during the protocol, while the power density in spindle range (11.1-15.0 Hz) and frequencies between 15.1-25.0 Hz showed strong circadian modulation (p<0.05, ANOVA 4-h intervals) which did not differ from baseline (p>0.5, ANOVA). Analysis of the time course of SWA and power density in the spindle range in the first 7 min of a NREM sleep episode confirmed these results, with no circadian modulation in SWA but circadian changes in the spindle range and a time course within the NREM sleep episode identical to baseline. The present data show that, in a contrast to SWA, HFA is not influenced by sleep homeostatic mechanisms and displays significant circadian modulation with endogenous origin. Supported by EU Grant LSHM-CT-2005-518189.