Resources
Authors & Affiliations
Marielena Sosa,Mark Plitt,Lisa Giocomo
Abstract
To learn and remember where to find reward in an environment, the brain’s spatial map must be updated by reward predictive information. This spatial memory process relies on the hippocampus. As reward locations are learned, hippocampal place cells shift or ‘remap’ their firing fields to overrepresent the reward location at the population level[1]. These shifts are thought to rely on dopaminergic input, which when manipulated can alter the hippocampal representation as well as the animal’s memory of reward[2-4]. Yet the dynamics of dopamine release in the hippocampus under normal learning conditions have yet to be described. Reward predictive dopamine signals have been established in other brain regions and provide a candidate for updating the hippocampal map, but it is unknown if such a signal is present in the hippocampus. Here, we provide insight into this question with simultaneous 2-photon (2P) imaging of calcium activity and a fluorescent dopamine sensor (GRABDA)[5] in hippocampal area CA1 as mice navigate virtual environments. We report the presence of two distinct dopamine dynamics: one reward-predictive signal which ramps up on approach to reward, and another which ramps down and putatively correlates with changes in running speed. These dopamine patterns shift to new reward locations in concert with changes in behavior and updates to the hippocampal spatial representation. To explore the organization of these dopamine signals, we performed multi-plane imaging across the dendritic arbor of CA1 pyramidal cells and found evidence that dopamine dynamics vary by anatomical depth, suggesting a drive for varied remapping properties across the CA1 sublayers[6]. In ongoing work, we are investigating the relationship of these dopamine signals to the remapping of CA1 place cells. These results establish the spatiotemporal dynamics of dopamine release in CA1 during spatial navigation for a remembered reward.