Final Report Summary - HIPECMEM (Memory-Related Information Processing in Neuronal Circuits of the Hippocampus and Entorhinal Cortex) This proposal investigated the circuit mechanism that underlies the spatial memory-related information processing in the hippocampus and the medial entorhinal cortex (MEC). Both of these regions are implicated in spatial memory and encode spatial information in neuronal activity patterns. To investigate these circuits process, we examined the neuronal code of spatial memory traces and their reactivation is sleep periods, which promotes memory consolidation. In addition, we examined alterations of neuronal connection weights during spatial learning. To investigate these research questions, we performed combined multi-electrode recordings and recorded from large populations of cells. Moreover, optogenetic techniques were used to transiently silence neuronal populations and examine their lasting effects in network function. Our previous work showed that many CA1 place cells change their firing as a result of spatial learning. When animals required to remember for a longer periods of time the location of goals, many place cells remapped their place fields to these goals, demonstrating that they encoded memories of goal locations. Now, we tested place cell expression dynamics during learning and found that old and new maps transiently alternated (flicker) during the early stages of learning, suggesting that new map formation is a competitive process during learning. We also examined firing of place cells in spatial working memory tasks on mazes in which animals had to remember locations only briefly. We found that place cells did not change their place fields; however, they altered their firing rate to encode working memories. Therefore, in the long-term goal memory condition place cells alter their place fields (global remapping), while, in short-term memory conditions, they only alter their within place field rate (rate remapping).To test the role of 200Hz, sharp wave/ripple (SWR)-associated reactivation in consolidation, we examined whether the reactivation of waking, place-related activity in sleep/rest period SWRs was required for the recall of new place maps. We found that optogenetically inhibiting hippocampal networks during SWRs did prevent the subsequent recall of new maps, suggesting that new memories of novel environments do not require SWR-associated reactivation for their consolidation. By contrast, when we optogenetically prevented place cells to be active during active waking periods at their already established place fields, this caused them to remap their place fields and their previous place fields did not return even after the inhibition was stopped. This suggests that waking activity is crucial for place map stability and even already established maps are labile when they are expressed during waking activity. To investigate connection changes in hippocampal circuits during learning, we examined the changes of connection weights between monosynaptic pyramidal cell-interneuron pairs during goal learning on the cheeseboard maze. We showed that such connection weight changes can occur exclusively during spatial learning without further changes taking place in sleep or subsequent memory probe trials. This work demonstrated that spatial learning includes wide-ranging circuit changes that involve both excitatory and inhibitory circuits. This project developed several novel research methodologies including the simultaneous recording of hippocampal and MEC cell assemblies and combining assembly recordings with optogenetic biasing of networks. We further developed 128-channel recordings, long-term (30h) radio telemetry recordings of assemblies and online assembly detection, enabling the selective, optogenetic inhibition of specific cell assemblies. These new techniques will help us to continue tackling novel questions about the circuit mechanism behind spatial memory formation.
