Symposium
New insights on the cortico-hippocampal dialogue underlying memory記憶を担う大脳皮質と海馬の協調の新展開
シンポジウム
New insights on the cortico-hippocampal dialogue underlying memory記憶を担う大脳皮質と海馬の協調の新展開 |
|---|
| 7月28日（日）8:45～9:09 第3会場（朱鷺メッセ 2F メインホールB）
4S03m-1
Episodic time coding in lateral entorhinal cortex
Albert Tsao（Tsao Albert）
Stanford University
The encoding of time and its binding to events are critical components for episodic memory, but how these processes are carried out within the hippocampal-entorhinal circuit is unclear. In this talk, I will discuss recent results suggesting a significant role for lateral entorhinal cortex (LEC) in generating temporal information. Recording from freely foraging rats, we find that temporal information is robustly encoded across time scales from seconds to hours within the overall population state in LEC. This temporal information was present both at the single-cell level in the form of ramping cells, as well as at the population level through a continuously-changing population activity state, and was unique to LEC. In addition, the evolution of LEC population activity was not a stationary process, but instead exhibited features suggesting a link to subjective time. Consistent with this link to subjective time, when animals' experiences were constrained by behavioral tasks to become similar across repeated trials, the encoding of temporal flow across trials was reduced while encoding of time relative to the start of trials was improved. The findings suggest that populations of LEC neurons represent time inherently through the encoding of experience. This representation may be integrated with spatial inputs from medial entorhinal cortex in the hippocampus, allowing the hippocampus to store a unified representation of what/where/when. | | 7月28日（日）9:09～9:33 第3会場（朱鷺メッセ 2F メインホールB）
4S03m-2
Prefrontal top-down control over memory encoding in the hippocampus
Kaori Takehara-Nishiuchi（Takehara-Nishiuchi Kaori）
Univ of Toronto
Experiences are remembered or forgotten. Seminal human imaging studies show that the magnitude of the activation of the prefrontal cortex and hippocampus is predictive of whether experiences are later remembered. Parallel animal studies show that the medial prefrontal cortex (mPFC) undergoes functional remodeling at the time of memory encoding and that the integrity of the mPFC is necessary for encoding-induced gene expression in the hippocampus. These findings challenge the traditional view that the hippocampus learns first, and the neocortex follows. Instead, the mPFC may regulate the mnemonic fate of event information by modulating memory encoding processes in the hippocampus. Consistent with this idea, we found that mPFC neuron ensembles abruptly changed firing patterns when naive rats underwent a transition from a neutral experience (incidental sequences of auditory stimuli) to a behaviorally relevant experience (pairings of the auditory stimuli with an aversive periorbital shock). The population-level change was driven by ~15% of neurons that increased their baseline firing rates upon the first shock presentation. Initially, these neurons responded only to the shock; however, within a few stimulus-shock parings, they developed reliable responses to the stimulus, suggesting a ultra-rapid association of the stimulus with the shock. In keeping with these correlational findings, we also found that artificially augmenting the prefrontal network was sufficient to enhance memory encoding. Specifically, when rats associated one of two neutral stimuli with an aversive stimulus, both behavioral responses and stimulus-evoked oscillatory activity in the mPFC differentiated the relevant stimulus (i.e., paired with the aversive stimulus) from the irrelevant stimulus (i.e., presented alone). Chemogenetic enhancement of mPFC excitatory neurons promoted the development of the selective neural responses and in turn, facilitated the formation of the differential stimulus association. Ongoing experiments examine whether two prefrontal efferent regions, the nucleus reuniens and lateral entorhinal cortex, are necessary for the memory enhancement via the mPFC activation. Overall, these data suggest that the dynamics of the mPFC neural ensembles during events provide a relevance-signaling mechanism through which the mPFC may exert executive control over the encoding of those events in the hippocampus. | | 7月28日（日）9:33～9:57 第3会場（朱鷺メッセ 2F メインホールB）
4S03m-3
Disruption of oligodendrogenesis impairs spatial memory consolidation in adult mice
Paul Frankland（Frankland Paul）
Neuroscience & Mental Health, Hospital for Sick Children
Activity-dependent changes in myelin patterning have been hypothesized to promote coordinated reactivation of neural patterns in distributed cortical regions that are important for the gradual consolidation of initially, hippocampus-dependent memories. Here we tested this hypothesis, and provide three lines of evidence that indicate that oligodendrogenesis critically contributes to the long-term consolidation of spatial memory in the adult brain. First, in adult WT mice, we found that oligodendrogenesis occurs after spatial learning. The generation of new oligodendrocytes occurred in the absence of any additional training, and was largely restricted to cortical and associated white matter regions that have been previously associated with long-term consolidation of spatial information. Second, preventing oligodendrogenesis following training impaired consolidation of spatial memories. Impairments were only observed when oligodendrogenesis was suppressed immediately following training, and not at more remote time points, revealing a critical post-training window for these effects. Third, oligodendrogenesis suppression reduced coupling of hippocampal sharp wave ripples and cortical spindles, two rhythmic oscillations that contribute to memory consolidation. Previous studies have focused on how neuronal changes, whether at the molecular, cellular or population level, contribute to memory consolidation. The current data support the idea that non-neuronal cell types contribute critically to consolidation by fine-tuning active circuits following learning. | | 7月28日（日）9:57～10:21 第3会場（朱鷺メッセ 2F メインホールB）
4S03m-4
Prefrontal cortical neurons reflect hippocampal non-local trajectory information during hippocampal replay and during theta sequences
David Foster（Foster David）1，Alice Berners-Lee（Berners-Lee Alice）1,2，Xiaojing Wu（Wu Xiaojing）2
1University of California, Berkeley
2Dept of Neuroscience, Johns Hopkins University School Of Medicine, Baltimore MD, USA
Using experience to guide future decisions is critical for adaptive behavior. It has been proposed that one way in which information from experience is organized in the brain is in the form of an internal model, sometimes also called a cognitive map. Such a model could allow animals to access spatially and temporally non-local information, in support of learning and decision-making. Growing evidence suggests that awake hippocampal replay can function as such a model, since it depicts non-local representations of future and past behavior. In particular, awake replays depict the future path to the goal in a spatial memory task, and depict prior behaviors when a reward is encountered. It is increasingly also clear that the characteristic LFP pattern in the hippocampus that is associated with replay, the sharp-wave ripple (SWR), occurs coincident with activity in many cortical areas, such as medial prefrontal cortex (mPFC), suggesting that hippocampal replay may propagate information about past or future trajectories to cortical areas directly involved in decision-making. However, it remains unknown whether or not cortical activity reflects the trajectory information carried by hippocampal replay. We recorded neurons in mPFC and hippocampus simultaneously while rats navigated a novel Y-maze, during early learning. We found that individual mPFC neurons were differentially modulated to replays depicting different arms of the maze. Indeed, the trajectory identity of the hippocampal replay could be decoded from the coincident mPFC firing alone. Interestingly, mPFC neurons did not show spatially selective activity patterns during running on the maze, despite the trajectory selectivity during hippocampal replay. However, during running, mPFC neurons exhibited trajectory selectivity during the non-local portions of hippocampal theta cycles (sometimes called theta sequences or sweeps), even while they were non-selective during the local portion of the theta cycle when hippocampal place cells were reporting current position. Thus, across both theta and SWRs, mPFC neurons appear to reflect hippocampal trajectory information but only during periods of non-local representation. These data may shed light on the dialogue between hippocampus and mPFC that may support model-based decision-making. | | 7月28日（日）10:21～10:45 第3会場（朱鷺メッセ 2F メインホールB）
4S03m-5
Physiological signature of memory age in the prefrontal-hippocampal circuit
Thomas J. Mchugh（Mchugh Thomas J.）
RIKEN Center for Brain Science, Lab for Circuit and Behavioral Physiology
Episodic memory formation depends on the hippocampus, however its long-term storage requires communication between the hippocmapal circuit and the prefrontal cortex. While much is known about the physiological coupling of these regions early in memory consolidation, how this time-dependent process alters their dynamic interactions during subsequent recall remains unknown. To address this question we performed longitudinal simultaneous electrophysiological recordings of local field potentials and single unit activity from the anterior cingulate cortex (ACC) and hippocampal CA1 region in mice during the recall of recent and remote contextual fear memories. We found that in contrast to recent memory, remote memory recall was accompanied by increased ACC-CA1 synchronization at multiple frequency bands, specifically theta and fast gamma. The augmented ACC-CA1 interaction was associated with strengthened coupling among distally spaced CA1 neurons, implying an ACC-driven organization of a sparse hippocampal code. Further, the robust shift in physiology we observed permitted a support vector machine classifier to accurately determine memory age based on the ACC-CA1 synchronization pattern alone, both within and between subjects. Our findings suggest that memory consolidation alters the physiological interactions between these regions in a stereotypical manner, supporting the Multiple Trace theory, and provide a novel biomarker for laboratory, and possibility clinical, studies examining memory impairments in age and disease. | | | |
|
|
