Deciphering the contribution of subicular networks to replay and memory
Lead Research Organisation:
UNIVERSITY COLLEGE LONDON
Department Name: Cell and Developmental Biology
Abstract
The brain stores and retrieves information about everyday experiences, forming memories that define who we are and how we behave-an ability that is affected in many dementias. The hippocampus is known to be a crucial actor involved in the transformation of labile memories into long-lasting ones, a process termed system-level memory consolidation. During quiescence, neurons in the hippocampus rapidly recapitulate sequences of previously experienced behavioural trajectories. These replays are considered to be a key mechanism supporting memory consolidation.
The subiculum is the main output of the hippocampus and sits at the crossroads of numerous communication pathways with cortical and subcortical structures. Thus, the subiculum might play a central role in mediating how replayed information interacts with extra-hippocampal structures. Surprisingly, the subiculum has been empirically and theoretically neglected, and its role in consolidation is still relatively unexplored. In this project, we will investigate how the subiculum supports memory formation using advanced techniques like two-photon calcium imaging, multisite electrophysiology, optical tagging of genetically defined cell types, and computational modelling.
An unusual feature of the subiculum is that it contains neurons that respond to extended environmental features, such as walls and barriers. Surprisingly, some of these neurons can maintain a trace of this activity even after the cue that elicited it is removed-these are known as trace responses. This discovery positions the subiculum as a potential 'buffer', retaining short-term memory traces of environmental features to enhance the possibility of their consolidation into cortical memory. In this study, we will explore the network properties and specific neural types that generate these traces, study the learning rules that establish them, and understand how they interact with hippocampal replay events.
In summary, these experiments will enhance our understanding of the subiculum's role in memory consolidation. Specifically, they will illuminate how the subiculum's unique cellular architecture and diversity contribute to trace activity and facilitate the consolidation of these traces into cortical networks.
The subiculum is the main output of the hippocampus and sits at the crossroads of numerous communication pathways with cortical and subcortical structures. Thus, the subiculum might play a central role in mediating how replayed information interacts with extra-hippocampal structures. Surprisingly, the subiculum has been empirically and theoretically neglected, and its role in consolidation is still relatively unexplored. In this project, we will investigate how the subiculum supports memory formation using advanced techniques like two-photon calcium imaging, multisite electrophysiology, optical tagging of genetically defined cell types, and computational modelling.
An unusual feature of the subiculum is that it contains neurons that respond to extended environmental features, such as walls and barriers. Surprisingly, some of these neurons can maintain a trace of this activity even after the cue that elicited it is removed-these are known as trace responses. This discovery positions the subiculum as a potential 'buffer', retaining short-term memory traces of environmental features to enhance the possibility of their consolidation into cortical memory. In this study, we will explore the network properties and specific neural types that generate these traces, study the learning rules that establish them, and understand how they interact with hippocampal replay events.
In summary, these experiments will enhance our understanding of the subiculum's role in memory consolidation. Specifically, they will illuminate how the subiculum's unique cellular architecture and diversity contribute to trace activity and facilitate the consolidation of these traces into cortical networks.
Technical Summary
The subiculum is the main pathway for hippocampal communication with cortical and subcortical targets. Its strategic position suggests it is key in mediating cortico-hippocampal interactions during system-level memory consolidation. Replay-rapid sequences of hippocampal neuron activity-is widely seen as a core mechanism supporting consolidation. Surprisingly, it is unclear whether the subiculum plays a role in this process, and if it exhibits replay. This is particularly perplexing as the subiculum's complex anatomical and computational characteristics suggest it has an active role in memory formation and consolidation. We aim to clarify this, focusing on two research streams.
First, we propose the subiculum forms short-term memory traces of new environmental features, enabling their consolidation into long-term memory. We will use two-photon imaging, electrophysiology, and computational modelling to identify these trace cells during active behaviour and replay. We predict that trace responses emerge from theta-scale interactions between recurrently connected neurons. We also hypothesise the subiculum exhibits two types of replay events: coherent with CA1 and independent. Finally, we propose that trace cells are preferentially engaged during replay, resulting in a gradual erosion of their trace responses.
Second, we propose that distinct populations of subicular pyramidal neurons-bursting and regular cells-are recruited differentially to form trace activity and route information out of the subiculum. To optically discriminate bursting cells, we will tag them using a genetically modified mouse line. We predict that subicular trace activity mostly occurs in bursting cells and that these are preferentially engaged in offline replay, which is believed to be primarily involved in consolidation.
First, we propose the subiculum forms short-term memory traces of new environmental features, enabling their consolidation into long-term memory. We will use two-photon imaging, electrophysiology, and computational modelling to identify these trace cells during active behaviour and replay. We predict that trace responses emerge from theta-scale interactions between recurrently connected neurons. We also hypothesise the subiculum exhibits two types of replay events: coherent with CA1 and independent. Finally, we propose that trace cells are preferentially engaged during replay, resulting in a gradual erosion of their trace responses.
Second, we propose that distinct populations of subicular pyramidal neurons-bursting and regular cells-are recruited differentially to form trace activity and route information out of the subiculum. To optically discriminate bursting cells, we will tag them using a genetically modified mouse line. We predict that subicular trace activity mostly occurs in bursting cells and that these are preferentially engaged in offline replay, which is believed to be primarily involved in consolidation.
