The basic functions of sleep are still unknown. Abnormal sleep patterns can manifest as a variety of disorders-sleep apnea, parasomnias, REM (rapid eye movement sleep) behavioral disorder (RBD), narcolepsy-many of which are influenced by heredity. There is an increasing focus on characterizing mouse behaviors for genetic and drug studies. However, discovering the genes responsible for sleep and related disorders requires time-consuming large-scale behavioral screening of phenotypes to correlate observed traits with genetics. Behavioral monitoring of mice is usually limited to actigraphic measurements such as video tracking, wheel-running, and photoelectric beam-breaking. Although many of these methods are noninvasive and have potential for high-throughput (HT) application, they monitor mainly locomotor activity without providing information about sleep-wake state and sleep architecture, which are important for investigating sleep disorders. The current gold standard for sleep analysis in mammals is electroencephalography (EEG) with electromyography (EMG). While EEG can be used to accurately determine sleep-wake state, it is invasive and resource-intensive (surgery, recovery, etc.), which limits its application in large scale genetic studies with rodents. EEG is therefore a critical barrier to the discovery of genes that promote sleep disorders. Signal Solutions, LLC, has developed a sensor cage environment for noninvasive, HT behavioral monitoring that is being used by many prominent research groups to identify genes responsible for different traits related to sleep and circadian rhythms. The system is based on analysis of the signal generated by a pressure- sensitive piezoelectric sensor attached to the cage floor, and can already discriminate sleep from wakefulness with high accuracy and track changes in respiratory effort when the animal is relatively inactive. The Sunderam Lab at the University of Kentucky has used Signal Solutions'"piezo" system to develop techniques and obtain preliminary data suggesting that pressure changes associated with respiratory effort may distinguish REM and non-REM (NREM) stages of sleep as verified by simultaneous EEG/EMG recordings.
The specific aims of this application are to determine whether the piezo system can noninvasively: 1. Discriminate sleep-wake state (sleep/wake, REM/NREM) and behavior within wake (e.g., quiet vs. active, high activity, feeding, grooming) at a level comparable to EEG/EMG by classifying piezo signal features;2. Identify outliers in a cohort and differentiate strains of mice with known sleep differences on the basis of specific sleep traits (percent time in each state, mean bout frequency and duration, sleep-onset REM);and 3. Develop the capability to apply and quantify responses to sensory stimulation for selective sleep restriction and startle reflex measurement. The purpose of this investigation is to integrate and test these additional capabilities in the piezo system. The envisioned end product is a sensor cage and software interface for high-throughput monitoring of sleep- wake state and behavior in small animals (e.g., KO mice, QTL analyses) to identify genetic factors responsible for sleep/circadian disorders as well as behavioral effects of pharmacological manipulation, sensory stimulation, or neural injury (e.g., traumatic brain injury, epilepsy). This system will be particularly advantageous for prescreening potentially interesting phenotypes, and reserving invasive EEG analysis for further confirmation. The current system for classifying sleep vs. wake is essentially as good as EEG/EMG;REM/NREM would be extremely valuable as a first pass screen. Medical targets of interest are sleep/circadian disorders, sleep apnea, obesity/diabetes, REM/NREM sleep deprivation, and stress, among others. Potential clients include academic research labs as well as industrial labs interested in behavioral monitoring on a large scale (e.g. drug screening), and upgrades to existing users.
Discovery of genes that play a role in sleep and circadian rhythm disorders requires extensive screening of behavior, usually in mice, preferably with invasive and resource-intensive brain signal (EEG) recordings to score sleep stage (REM, NREM) and wake behavior. The goal of this research is to develop and validate a methodology for using a noninvasive, pressure-sensitive piezoelectric (piezo) sensor platform to distinguish different stages of sleep and behavior without the need for EEG. This would make large-scale behavioral screening feasible and limit the need for EEG verification to only the most interesting phenotypes thus identified. Beyond the ability to stage sleep and behavior a novel method is proposed here to selectively restrict sleep using sensory stimulation whenever a particular event or stage of sleep is detected from the piezo signal. This method, if validated, will be helped us analyze the effects of restriction of total sleep or particular stages of sleep on health and performance. It will also be useful for assessing how reflexive brain responses are altered by injury (e.g., TBI) or neuropsychiatric disorders (e.g., anxiety, depression) or other diseases.