Explorando a corticalização das memórias: marcadores eletrofisiológicos da consolidação da memória na tarefa do Labirinto de Barnes

Abstract

Communication between the hippocampus and associative neocortical regions is indispensable for the consolidation of declarative memories during sleep. In particular, it is known that communication between the hippocampus and the medial prefrontal cortex (mPFC) is extremely important for the consolidation processes of memories related to spatial goals. Specifically, the Barnes Maze (BM) task, with multiple acquisition sessions, allows the individual activity of these regions as well as their joint activity to be evaluated during sleep following learning over several days. However, knowledge about the electrophysiological markers of hippocampus-mPFC communication-dependent memory consolidation during sleep is still limited. Therefore, this study aims to investigate the role of communication between the hippocampus and mPFC during sleep in the consolidation of memory related to learning in the BM and the electrophysiological markers that evidence this process. To achieve this objective, male Wistar rats had recording electrodes implanted in CA1 and the mPFC. Subsequently, these animals underwent the Barnes Maze task acquisition protocol over 4 days and the test protocol on the fifth day. Additionally, the local field potential (LFP) of CA1 and mPFC was recorded during sleep following learning. DeepLabCut, an open-source software for animal pose estimation using deep neural networks, was used to track relevant parts of the animal's body and the maze frame-by-frame. Subsequently, a set of Python routines was used to extract the most relevant behavioral parameters from each acquisition and test session in the BM. Furthermore, from the LFP records, the effect of learning in the BM on (1) sleep architecture; (2) oscillatory activity of CA1 and mPFC; (3) synchrony between CA1 and mPFC; (4) interaction between different hippocampal rhythms; (5) occurrence of hippocampal sharp-wave ripples (SWRs), delta waves, and mPFC spindles during sleep was evaluated. Additionally, the animals' performance was correlated with the variation of the parameters assessed in the sleep of the first acquisition day compared to baseline recording. The animals learned the task and effectively consolidated the memory of the BM escape location, as evidenced by the increased use of the direct strategy over the acquisition days. Also, by the longer time spent in the maze's target quadrant on the test day. LFP records showed that learning in the BM generates profound changes in hippocampal and neocortical networks. Specific changes in sleep architecture were observed, such as an increase in the number of awake bouts on the first acquisition day and an increase in time spent in NREM and REM sleep from the second hour of recording only on the BM acquisition days. Moreover, significant changes in spectral characteristics of the signal were observed, with an increase in the mean power of the delta band in CA1 and mPFC on the BM acquisition days. There was also a gradual decrease over the recording hours. No changes were observed in phase coherence or spectral coherence between CA1 and mPFC during sleep. However, the learning process produced changes in hippocampal and neocortical networks, increasing not only the occurrence of delta waves in the mPFC and hippocampal SWRs, but also the joint occurrence of these events. Finally, an increase in coupling between the phase of delta oscillations and the amplitude of high-frequency oscillations (90-140 Hz) during NREM sleep on the BM acquisition days was observed. In particular, it was demonstrated that slow wave activity, SWR duration, delta wave occurrence, spindles, and delta waves coupled to SWRs may be possible electrophysiological markers of initial memory consolidation related to learning in the BM. Therefore, this study corroborates the relevance of SWRs, spindles, and delta waves in processes relevant to memory consolidation. It also provides new evidence on regional changes and communication between CA1 and mPFC regions as a consequence of learning in the BM.