Sparse and Specific Coding during Information Transmission between Co-cultured Dentate Gyrus and CA3 Hippocampal...

Area of Science:

• Neuroscience
• Computational Neuroscience
• Cellular Neuroscience

Background:

• The hippocampus plays a crucial role in memory formation and spatial navigation.
• Understanding information processing in hippocampal sub-regions like the dentate gyrus (DG) and CA3 is key to deciphering neural computation.
• Engineered neural networks offer a platform to study specific circuit dynamics and information flow.

Purpose of the Study:

• To investigate the encoding and decoding of stimulus information in a co-cultured DG-CA3 hippocampal network model.
• To test the hypothesis that feed-forward DG to CA3 information flow is more accurate than reverse CA3 to DG flow.
• To characterize the neural code, including sparseness and uniqueness, within these hippocampal sub-regions.

Main Methods:

• Co-culturing rat primary DG and CA3 neurons in a two-chamber device with micro-tunnel axonal connectivity.
• Utilizing a micro-electrode array (MEA) for stimulation and recording across an 8x8 grid.
• Applying paired-pulse stimulation and employing Support Vector Machine (SVM) learning algorithms for decoding stimulus site information.

Main Results:

• Stimulation in DG evoked sparse and specific responses in CA3, indicating a selective neural code.
• CA3-DG networks exhibited less sparse coding compared to DG-DG and CA3-CA3 control networks.
• Decoding accuracy of stimulation sites in DG from CA3 recordings was 43%, significantly higher than the reverse (DG from CA3).

Conclusions:

• The engineered DG-CA3 co-culture model successfully recapitulates aspects of the in vivo hippocampal network.
• Sparse and specific neural responses in CA3 are selectively evoked by DG stimulation.
• Information decoding is more accurate in the feed-forward DG to CA3 direction, supporting its role in hippocampal information processing.