Neurochemical mechanisms of sleep-dependent memory consolidation

Abstract

The beneficial influence of sleep on memory has received considerable support during the last decade. The most widely accepted mechanism for the sleep-dependent strengthening of memories acquired during preceding wakefulness is that of their reactivation during slow wave sleep. The present thesis focused on pharmacological manipulations of sleep-dependent memory consolidation to elucidate neurochemical mechanisms that translate this reactivation into plastic changes. To this end, participants learned memory tasks before a retention interval, during which an active agent was administered. The most abundant excitatory neurotransmitter in the human brain glutamate participates in many forms of plasticity. The best studied form of plasticity is that of the glutamatergic synapse and relies on AMPA and NMDA receptor interaction. However, it is unclear if glutamatergic neurotransmission mediates the sleep-dependent consolidation of declarative memories. Hence, altering the action of these receptors during sleep, promises insights into the neurochemical mechanisms of plasticity that lead to sleep-dependent memory consolidation. Using AMPA and NMDA receptor blockers during sleep did not influence the consolidation of a declarative word pair associates task. Participant’s performance on the same task, however, was significantly increased, if d-cycloserine, a NMDA receptor co-agonist, was given during sleep. This result indicates that NMDA receptor mediated plasticity is important for sleep-dependent declarative memory consolidation and that the processes involved may differ from those observed during wakefulness. Dopamine is an important neuromodulator that can influence memory strength by facilitating synaptic plasticity. Activity of dopaminergic reward circuitry leads to better learning of rewarding information. The goal of the second study was to elucidate if dopaminergic circuitry is also involved during the reactivation of reward memory during sleep. Participants had to learn pictures, for which they were promised a high or a low reward at retrieval. Giving pramipexole, a d2-like receptor agonist, during the sleep retention interval wiped out the high over low reward benefit observed under placebo. Importantly, this effect was independent of encoding depth and thus speaks for reactivation of dopaminergic reward circuitry during sleep influencing the fate of memory encoded during prior wakefulness. Inhibitory neurotransmission in most cases involves the neurotransmitter GABA. GABA has also been shown to be involved in switching between states of arousal and sleep. It is important for the generation of the slow frequency EEG oscillations characteristic of slow wave sleep that have been shown to benefit sleep-dependent consolidation of declarative memory. In the third study administration of the GABA reuptake inhibitor tiagabine during retention sleep was used to manipulate this generating mechanism of slow wave sleep with the aim of boosting declarative memory retention. As expected, this manipulation highly increased the amount of slow wave sleep and slow oscillations that participants displayed, but the associated sleep spindle activity was dampened. As declarative memory was unexpectedly not enhanced by this treatment, it seems probable that effective slow wave sleep must also be accompanied by sleep spindle activity phase-locked to slow oscillations. Together these studies demonstrate how the pharmacological approach can yield information about the neurochemical underpinnings of sleep-dependent memory consolidation and identified neurotransmitter systems that are involved in this process. Future pharmacological experiments, however, also in animal models, will have to specify the mechanisms we are only beginning to understand.