Modeling lactate diffusion in gray matter in the context of the astrocyte-neuron lactate-shuttle

Motivation: Astrocyte-neuron lactate shuttle hypothesis (ANLS) remains fiercely debated due to the difficulty to assess lactate compartmentation non-invasively. Goal(s): Using time-dependent DW-MRS and a new, three-compartment model, we propose to assess the neuronal, astrocytic and extracellular fraction of lactate. Approach: A three-compartment model is built with parameters describing intra-neuronal and intra-astrocytic lactate diffusion derived from NAA and myo-inositol diffusion, respectively, and an extracellular compartment considered Gaussian. Results: Using this model, neuronal lactate fraction is estimated at ~4%, astrocytic fraction at ~66% and extracellular fraction at ~40%, suggesting a concentration gradient from astrocytes to neurons compatible with ANLS. Impact: Three-compartment modeling of lactate diffusion is a promising tool for assessing lactate compartmentation in gray matter. It yields lactate fraction in astrocyte, neurons and extracellular space compatible with ANLS, but its use remains difficult when noise is large.