Fully automated sleep deprivation in rats using air puffs

Abstract

Full sleep deprivation via gentle handling is time-consuming and can result in increased personnel costs. The objective of this project was to develop an automated air puff sleep deprivation system using real-time state detection based on our previously published off-line software for automated state scoring. Data were acquired from male Sprague–Dawley rats online in real- time in 2 s epochs which are scored using nuchal EMG and a cortical EEG signals. Core blood was collected from sleep deprived and control rats after 6 h of air puff sleep deprivation during the light cycle or subjected to the same 6 h sequence of air puffs during the dark cycle to account for circadian fluctuations in corticosterone levels. Serum from each blood sample was isolated using a centrifuge and corticosterone concentration was measured using a radio immunoassay according to established methods. Analysis of corticosterone levels in serum showed no difference between the rats exposed to air puffs and their respective controls, suggesting that the air puffs are not stressful to the rats. Corticosterone levels in the dark cycle were higher than in the light cycle in accordance with published studies. Manual state scoring using the EMG and cortical EEG in 10 s epochs confirmed that the system completely deprived cortical sleep. Further investigation using a male Long Evans rat implanted with multi-tetrode micro- drive in the hippocampus and cortical EEG screw electrodes also showed full sleep deprivation according cortical leads, consistent with the Sprague- Dawley rats. However, deep hippocampal EEG showed that local non-REM sleep occurred in the hippocampus approximately 10% of the sleep deprivation period. When not sleep deprived, no significant difference in the amount of non-REM sleep occurred between cortically and hippocampally scored EEG (Student’s t -test; p = 0.92). The results show that the automated air puff sleep deprivation system completely prevents sleep when based on cortical EEG and is no more stressful to the rat than being housed in a recording chamber. This system can thus be used as an automated alternative to gentle handling in experiments that require full deprivation of sleep. Local sleep in the hippocampus, despite full sleep deprivation in the cortex, may have implications for compensatory mechanisms of memory consolidation during sleep deprivation and will be investigated in future cognitive studies. Research funded by: The Dept. of Anesthesiology, University of Michigan Medical School and the National Institutes of Health (NIMH 1 R01 MH60670–11).