Abstract
Temporal lobe epilepsy (TLE) is a devastating disease, often pharmacoresistant and with a high prevalence of 1% worldwide. There are few disease-modifying therapies; thus, prevention has become a health priority. The overarching goal of this research project is to highlight the system's dynamics at different stages before TLE onset to identify an early shift in network dynamics trajectory towards disease onset. Researchers often investigate collective brain activity by tracking dynamical interactions of the signal recorded at multiple sites. However, these interactions are usually only computed between pairs of brain regions, at the risk of missing simultaneous interactions of three or more areas, an aspect that is crucial in a networked disease such as TLE. We thus propose to track, on a rich dataset of electrophysiological brain signals recorded within the TL of adult male Wistar Han rats, the formation and dissolution of high-order informational multiplets in time during distinct natural behaviors in an animal model of TLE. We identified the informational content of the multiplets as synergistic or redundant. Results identified an early transition of synergistic and redundant multiplets ahead of TLE onset with the predominant involvement of four TL brain regions in generating theta (4-12 Hz) activity. This shift has been shown predominantly during exploration, a theta-dependent behavior, less during rest and sleep. This specific change suggests a shift in communication from an integrated to a segregated network toward TLE onset.
Significance Statement Temporal lobe epilepsy (TLE) is a prevalent network disorder that is often pharmacoresistant; therefore, TLE prevention is critical. This research article identifies early signs of TLE by studying collective dynamics within the TL using an information decomposition technique during natural behaviors. This computational method, beyond pairwise interactions, may thus need to be used to identify early biomarkers for TLE onset.
Footnotes
We have no conflict of interest to declare.
We thank the Institute for Molecular and Behavioral Neuroscience, University of Cologne, for support to LC, also Dr. Christophe Bernard for his valuable comments on a previous version of the manuscript and the use of the experimental data acquired in his lab at the Institut de Neurosciences des Systèmes (INS), Marseille, France. We also thank Brechje Buskens (MSc at Radboud University Medical Center, Nijmegen, The Netherlands) for her contribution to obtaining preliminary results, which are not fully reflected in this publication but have been used to initiate this work. We also thank the reviewers for their pertinent comments on an earlier manuscript version.
This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license, which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.
