Back

Morpho-electric and computational properties of three types of human hippocampal CA1 pyramidal neurons

Mertens, E.; Leibner, Y.; Pie, J.; Galakhova, A.; Waleboer, F.; Meijer, J.; Heistek, T.; Wilbers, R.; Heyer, D.; Goriounova, N.; Idema, S.; Verhoog, M.; Kalmbach, B.; Lee, B.; Gwinn, R.; Lein, E.; Aronica, E.; Ting, J.; Mansvelder, H.; Segev, I.; de Kock, C.

2023-10-06 neuroscience
10.1101/2023.10.05.561029 bioRxiv
Show abstract

Hippocampal pyramidal neuron activity underlies episodic memory and spatial navigation. Although extensively studied in rodents, extremely little is known about human hippocampal pyramidal neurons, even though human hippocampus underwent strong evolutionary reorganization and shows lower theta rhythm frequencies. To test whether biophysical and computational properties of human CA1 pyramidal neurons can explain observed rhythms, we map the morpho-electric and computational properties of individual CA1 pyramidal neurons in human, non-pathological hippocampal slices from neurosurgery. Human CA1 pyramidal neurons have extensive dendrites and resonate at 2.9 Hz, optimally tuned to human theta frequencies. Morphological and biophysical properties reveal three cell types with distinct dendrite bifurcations and physiology. Data-driven biophysical models show that human CA1 pyramidal neurons use i) computationally independent dendritic compartments, ii) preferential routing of electrical activity towards soma or dendritic tree and iii) non-linear input-output transformations. Across cell types, morpho-electric properties consistently increase computational richness in human CA1 pyramidal neurons. TeaserHuman CA1 pyramidal neurons have large and intricate morphologies, which translates to complex computational properties.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.