Skip to main content

The Brainstem-Hippocampus Network Uncovered: Dynamics, Reactivation and Memory Consolidation

Periodic Reporting for period 1 - DREAM (The Brainstem-Hippocampus Network Uncovered: Dynamics, Reactivation and Memory Consolidation)

Reporting period: 2019-08-01 to 2021-07-31

Sleep has various physiological functions, including the consolidation of memories. Current theories establish that memories are transiently stored in the hippocampus and transferred to the neocortex for long-term storage during sleep. In particular, during offline brain states, the hippocampus enables the reactivation of training trials through the spontaneous retrieval of previous memories. These processes are associated with plastic changes in distinct brain circuits, heralded by different types of macroscopic electrical activities that occur upon changes in the neuromodulatory activity of the brainstem. The brainstem prompts periods of both enduring and transient changes of neuronal excitability that affect the activity of other sub-structures in a precise manner. Thus, brainstem activity is likely critical for long-range coordination of several brain structures underlying memory formation. However, what specific neural mechanisms allow these subcortical regions to participate in memory formation?

The aim of this project is to uncover the role of different sleep stages in memory formation and long-term consolidation, and to elucidate the functional role of sleep-associated subcortical coordination events in these processes. To achieve this aim, the first objective is to identify memory reactivations in the hippocampus across sleep stages, and to quantify how these reactivations relate to specific oscillatory events in pontine and thalamic nuclei. The second objective is to identify specific neural mechanisms that trigger different hippocampal synchrony regimes specifically mediated by the brainstem activity during wakefulness and sleep. The third objective is to selectively interrupt the activity of pontine nuclei in order to determine their causal role in regulating hippocampal circuits, and the behavioural performance of the animals. Finally, the fourth objective is to elucidate through computational simulations the microcircuit mechanisms that mediate the interactions between the pontine nuclei, thalamic nuclei and the hippocampus, possibly underlying systems and synaptic consolidation.
1. In vivo electrophysiological recordings, behavioural apparatus and neural data analysis.
Wild-type Long Evans (LE) rats (N = 5 animals) were chronically implanted with recording micro-drives, incorporated with movable recording electrodes and with custom-made bipolar electrodes. Implantations were performed under deep anaesthesia, in addition to analgesia provided pre- and post-surgically. Following the animals’ recovery period (ca. one week), the animals were exposed to a behavioural apparatus, consisting of a cheeseboard maze and a sleeping cage. In parallel, during the same habituation period, individual electrodes were lowered manually towards the target regions (hippocampus, thalamus and brainstem).

Each animal was trained to collect food rewards on the maze on the basis of a reference memory task. Besides, we also focused on probing the neural responses of the pontine, thalamic and hippocampal circuits during periods of natural sleep. To this end, each animal (N = 4) was recorded for a period of 3 to 5 hours of daily natural sleep. These experiments were designed and performed to address the first two objectives of this research.

States of vigilance corresponding to waking, non-rapid eye movement (NREM) sleep, and REM sleep could be reliably identified from the hippocampal neuronal activity of each animal, in addition to the animals’ movement in the sleeping cage. Ponto-geniculo-occipital (PGO) activity (PGO waves) was characterized by the spontaneous appearance of negative deflections in the ponto-thalamic recordings in naturally sleeping rodents. Interestingly, PGO waves occurred both in NREM and REM sleep, and specifically coupled to REM-associated theta waves and NREM high-frequency ripples. When analysed separately, these two types of PGO episodes were associated with opposite profiles of brainstem population firing, at the same time, pointing to a significant putative causal influence of the brainstem over thalamic and hippocampal circuits. Crucially, learning of the reference memory task described above resulted in an increase in the coupling between PGO waves and hippocampal events, observed during post-learning sleep periods.

2. In vivo electrophysiological recordings combined with optogenetics in animals expressing light-sensitive ion channels in the brainstem’s parabrachial nucleus (PbN).
The aforementioned experiments have allowed us to partly elucidate the role of brainstem neural circuits in learning and memory consolidation. We next will test the hypothesis that selective interruption of the brainstem-hippocampus synchronizing pathway –by way of silencing pontine nuclei– during sleep prevents the formation of memories. To this end, similar behavioural and recording paradigms have to be performed in rats whose brainstem cells express the light-sensitive ion channel channelrhodopsin-2 (ChR2), thus conferring labeling of, and optical control over, brainstem cell populations through delivering light over implanted optic fibres. Given the level of technical difficulty of the experiments, together with the high time and technical demand of our previous experiments, objectives 3 and 4 of this action have not been yet accomplished in the current state of our research. Crucially, our latest pilot experiments show, for the first time, that brainstem cells can indeed selectively express the aforementioned light-sensitive ion channels in wild-type rats. These results indicate that we can indeed gain optical control over such brain nuclei during future in vivo experimental preparations.

I will disseminate the results of this investigation into three research articles to be published in leading international peer-reviewed journals (e.g. Nature, Science, Nat Comm, Neuron, eLife, PNAS). In addition, a review article about systems and synaptic memory consolidation in the mammalian brain integrating evidence from theoretical, molecular and systems neuroscience subfields (to be published in Nat Rev Neurosci, or similar) is also in preparation. All articles will target a broad scientific audience. In order to maximise the coverage of the dissemination, all articles will be published as Gold Open Access under a Creative Commons Attribution License (CC BY).
The results of this investigation indicate that the control of hippocampal ensembles by PGO waves might be a phylogenetically well-conserved neural mechanism. Our results indicate that these episodes may correspond to windows for promoting hippocampal-cortical communication and plasticity during NREM and REM sleep, likely promoting memory consolidation and synaptic homeostasis. In addition, our data open, for the first time, the possibility to determine the correlates, neuronal mechanisms and physiological significance of the PGO phenomenon in memory formation and long-term consolidation in rodents. This knowledge will contribute to the understanding of the physiological role of sleep and will uncover specific neural mechanisms underlying their role in the consolidation of memories. The aforesaid topics bear critical relevance in clinical settings and clinically-applied research; specifically, for establishing potential biomarkers and interventional targets for memory-related pathologies, sleep and sleep-associated neurological disorders.
Concurrent brainstem-thalamic-hippocampal recordings in behaving animals