The brain creates internal 'maps' to help us navigate and learn from our surroundings, but how these maps form remains a challenge to understand. Now, a study led by Liset M. de la Prida at the Cajal Neurosciences Center (CNC-CSIC) in Madrid, in collaboration with Imperial College London, offers a fresh perspective on how spatial and experiential information is encoded in the hippocampus, a key brain region for navigation and memory.
The study published today in Neuron, reveals that two types of hippocampal neurons work complementarily in mice to generate spatial maps that are more sophisticated and adaptable than previously thought. The findings open new avenues for understanding how the brain processes and represents spatial information, and could have implications for the future development of treatments for neurological disorders related to memory and orientation, such as in Alzheimer's disease.
The researchers found that two distinct subpopulations of pyramidal neurons, classified as superficial and deep based on their location in the hippocampus, respond differently during movement and rotation, allowing the brain to generate complementary geometric representations.
The study shows that deep pyramidal neurons respond to local changes, such as the position of furniture in a room. In contrast, superficial pyramidal neurons maintain a more stable spatial representation, focused on global features like the orientation of windows or doors. This stability is crucial for maintaining a consistent reference of the environment.
This study builds on research recognized with the 2014 Nobel Prize awarded to May-Britt and Edvard Moser, along with John O'Keefe, for their discoveries on place and grid cells, which form the foundation of the brain's positioning system. Liset M. de la Prida's findings expand on these discoveries, offering a deeper understanding of how the hippocampus encodes and processes spatial information by integrating different frames of reference.
Space, speed and direction
To study these representations, the team used simple maze-like corridors with visual and tactile cues, where mice could run back and forth. They discovered that deep-layer neurons were more tuned to space, speed, and the direction of movement than superficial-layer neurons. Interestingly, deep neurons responded to nearby landmarks, whereas superficial neurons were more attuned to visual cues in the broader room environment.
"Hippocampal neurons create abstract spatial representations that act as a map, enabling orientation and memory of past experiences. Until now, it was unclear how different neuron types contributed to these abstract maps, as these representations emerge from collective activity. It's like trying to determine which musicians in an orchestra are responsible for rhythm and which for melody—while all of them contribute, some play key roles in the final outcome," says Liset M. de la Prida.
The breakthrough was made possible by advanced imaging techniques that allow the simultaneous visualization of hundreds of neurons of each type. "We used two different-colored sensors to track the activity of both superficial and deep cells in real time," explains Juan Pablo Quintanilla, the co-author responsible for these experiments. This microendoscopic cellular imaging technique was established for the first time in Spain at the Cajal Neurosciences Center of the Spanish National Research Council (CSIC).
Maps updated in real time
Another key innovation of the study was the use of topological methods—a branch of mathematics that studies the properties of geometric objects—to decode the structure of these abstract neuronal maps. As mice explored the corridors, the hippocampal maps formed by the activity of hundreds of neurons from each subpopulation took the shape of three-dimensional rings.
When the environment changes—such as after rearranging furniture in a room—deep and superficial pyramidal neurons react differently. This allows the brain to update its spatial map while preserving coherent and flexible information about location and orientation despite external changes.
These different spatial representations (local versus global) coexist in parallel within the hippocampus. This ability to maintain multiple reference frames simultaneously is a remarkable feature of cognitive maps.
"The maps generated by these two neuronal subpopulations are interwoven, representing both global (e.g., the room) and local (e.g., the furniture) information," explains Julio Esparza, a biomedical engineer responsible for the topological analyses.
To test this flexibility, the researchers used chemogenetics to temporarily "silence" specific cell types. "We found that we could rotate the internal spatial maps and reorient the rings by selectively silencing superficial or deep neurons," Esparza explains.
The brain's ability to create mental maps can be used to improve memory, as in the 'memory palace' technique. This method helps people remember information by imagining familiar places, like their home or neighborhood, and mentally placing ideas in different spots to make them easier to recall.
The findings from the CSIC team open new avenues for understanding how the brain processes and represents spatial information and could have implications for developing future treatments for memory-related neurological disorders, such as Alzheimer's disease.
