Mapping GABA+/Glx in experimental temporal lobe epilepsy using edited-MRSI at 9.4T

Mesial temporal lobe epilepsy (MTLE) is a common drug-resistant epilepsy, often requiring surgery to remove the epileptogenic zone (EZ). The mean efficiency of the surgery is 50-70 %, so the accurate spatial localization of the EZ remains a challenge. In a previous study from Hamelin et al. (2021), gamma-aminobutyric acid (GABA), the main inhibitory neurotransmitter, was shown to be highly increased in the EZ of a MTLE mouse model, with a concomitant decrease of Glx (Glutamate + glutamine). The authors proposed the GABA/Glx ratio as a potential specific biomarker of the EZ. As it is the only way to measure GABA and Glx non-invasively in vivo, we propose to use Magnetic Resonance Spectroscopy (MRS) methods to improve the non-invasive localization of the EZ. Our study introduces an original GABA-edited Magnetic Resonance Spectroscopic Imaging (MRSI) method, i.e. the MEGA-LASER CSI, and the dedicated data processing pipeline, to map GABA and Glx in pre-clinical settings. The MEGA-LASER CSI sequence was validated using multi-compartment phantoms with varying GABA concentrations, demonstrating a high accuracy to spatially discriminate the different compartments. In vivo experiments revealed a significant increase of the GABA+/Glx ratio in the EZ of epileptic animals (n=30) compared to SHAM ones (n=15), strongly correlated with ex vivo data. Preliminary immunohistochemistry experiments revealed astrocytic localization of GAD65 in the EZ, suggesting a shift in GABA synthesis from nerve endings to astrocytes. Phantom and in vivo experiments prove that our original workflow for GABA and Glx mapping is suitable for use in neuroscience and pre-clinical applications.