Neuronal morphologies built for reliable physiology in a rhythmic motor circuit.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Systems Neuroscience

Background:

• Highly-branched neuronal structures are often assumed to perform compartmentalized computations.
• Previous work demonstrated the Gastric Mill (GM) neuron in the crustacean stomatogastric ganglion (STG) acts as a single electrotonic compartment despite extensive branching.

Purpose of the Study:

• To investigate if compact electrotonic architecture and linear voltage integration are generalizable to other STG neuron types.
• To elucidate the morphological and biophysical basis for these computational properties in STG neurons.

Main Methods:

• Simulations of 720 cable models with diverse geometries and passive properties.
• Analysis of neurite geometry, including tapering diameters.
• Broad parameter search to identify solutions for electrotonic properties and computational strategies.

Main Results:

• Compact electrotonic architecture and linear voltage integration are generalizable to other STG neuron types.
• These neurons exhibit direction-insensitive voltage integration, indicating pooled synaptic input.
• Neurite geometry, specifically tapering from 10-20 µm to <2 µm, underlies compact electrotonus, linear integration, and directional insensitivity.
• Multiple morphological and biophysical solutions exist for achieving varying degrees of electrotonic decrement and computational strategies.

Conclusions:

• STG neurons possess a compact electrotonic architecture that supports linear integration of synaptic inputs.
• Neurite geometry is a key determinant of passive electrotonic properties and computational strategies in these neurons.
• The findings challenge assumptions about compartmentalization in highly-branched neurons and highlight the role of morphology in neural computation.