Closed-loop auditory modulation of slow waves: development and application in healthy and traumatic brain injury rats

Closed-loop auditory stimulation (CLAS) of slow waves, characterized by low degree of invasiveness and high target specificity, is established as a highly promising tool in both basic and clinical sleep research. It constitutes a first of a kind alternative to the burdened pharmacological interventions currently dominating sleep medicine. For that reason, the application of CLAS in animal models is essential for a mechanistic understanding and technical optimization. Firstly, we report the development and characterization of what we believe is the first rodent model of phase-locked CLAS during slow wave sleep. Thus, we tested the effects of up- and down phase CLAS on proportions and spectral characteristics of sleep in healthy rats during the training-phase of the single pellet-reaching task. Without affecting 24-h sleep-wake behaviour, CLAS specifically altered SWA and sigma power over a chronic period of stimulation. Down-phase CLAS exerted a detrimental effect on engagement and success rate in this behavioural test. Therefore, modulation of SWA, altering the nesting of memory-consolidating spindle activity and, additionally, regulating intrinsic motivational states, appears as responsible for modified learning performance in our novel rodent model of CLAS. Secondly, we implemented this technique in a TBI rat model. In this context, elevated SWA is believed to be essential for efficient intracellular and extracellular clearance of trauma-related noxious subproducts, such as amyloid-β precursor protein (APP). Our past work showed that both sleep induction (GHB administration) and sleep restriction (SR) in TBI rats promoted enhanced SWA, resulting in increased APP clearance and improved cognition. Although promising, these sleep modulatory procedures have several drawbacks, including lack of specificity. Therefore, we used CLAS to precisely modulate SWA immediately after trauma. Up phase CLAS enhanced SWA, reduced APP+ axonal bulbs in the corpus callosum and improved cognitive ability in TBI animals. Conversely, targeting the down-phase of slow waves down-regulated slow oscillatory activity, and did not offer any beneficial effect over TBI-sequalae. Overall, we attempt to address a fundamental gap in technology — i.e. the implementation of CLAS to the rodent brain — that has so far hindered our understanding of the mechanistic link of auditory-modulated slow waves with learning and memory, as well as other brain processes strongly influenced by slow-wave sleep, such as the glymphatic function. At the very same time, given that it allows for specific and physiological modulation of sleep depth, this rodent technology represents a unique approach to explore the open questions of sleep functions. Our results provide proof of principle evidence that phase-targeted CLAS of slow waves is effective, safe and stable over chronic experimental periods, and offer pre-clinical indication of potential non pharmacologic slow-wave modulatory clinical applicability with therapeutic implications