Abstracts

Rationale: Sound-induced seizures are common models of epilepsy seen in a variety of species including rat, mouse, and hamster. Two strains of audiogenic-susceptible rats, the Genetically Epilepsy Prone Rat (GEPR3) and the Wistar Audiogenic Rat (WAR) have been widely used to examine the circuitry underlying generalized tonic-clonic seizures. The GEPR-3 strain was developed in the US from Sprague-Dawley (SD) rats, whereas the WAR strain was developed in Brazil from Wistar (WIS) rats. These strains show similarities at the level of seizure phenotype and general observation. However, despite decades of use, no work has directly compared them. To address this gap in knowledge, we examined the structure of GEPR and WAR brains as compared to their control strains using voxel-based morphometry (VBM) and diffusion tensor imaging (DTI). Methods: 60 day old, male GEPR and SD rats (Georgetown University) and WAR or WIS rats (Ribeirão Preto School of Medicine - University of São Paulo) were used. Rats were anesthetized, perfused, and brains post-fixed in the skull for 48h. Brains were then removed and stored in phosphate buffer prior to embedding in 3% agar at Georgetown. MRI (resolution of 11.7µm) was performed in a 7-T Bruker spectrometer with a 40-mm transmit-recieve volume coil. For VBM, a 3D T2-weighted RARE sequence was used. For DTI, a DTI-EPI SpinEcho sequence (30 diffusion directions) was used. VBM processing was performed to extract (MIPAV), resize, resample, and affine transform (FSL), and segment, smooth and normalize (SPM8). Statistical maps were analyzed using CAT12 in SPM12. A P < 0.001 height threshold with a cluster-wise FWE correction at P < 0.05 was used to determine statistical significance. DSI Studio was used for DTI processing, after which fractional anisotropy (FA) maps were exported for manual tracing by an observer blind to scan identify. Results: GEPR vs SD: GEPRs showed increased volume in superior colliculus (SC), periaquaductal grey (PAG), corpus callosum (CC), and claustrum. GEPRs showed decreased volume in fornix (Fx), CC, striatum, rhinal cortex, piriform cortex, and cerebellar vermis (Cb). WAR vs WIS: WARs showed increased volume in septum/fornix, striatum, amygdala/posterior piriform cortex. WARs showed decreased volume in the Cb, CC, and lateral geniculate. GEPR vs. WAR: volumes differed in primary motor and somatosensory cortex, inferior colliculus (IC), and hypothalamus. Only one region (Cb) differed in both strains as compared to their respective controls, an effect that was confirmed by hand-tracing (P Conclusions: Many of the brain areas we identified play integral roles in either the forebrain (piriform cortex, Fx, amygdala) or brainstem (IC, SC, PAG) seizure networks. Despite the similarities in the seizure phenotype between the GEPR and WAR strains, there are substantial differences in their brain structures. The area of overlap (i.e., Cb) may be of interest for further evaluation of the pathophysiology of audiogenic seizures in the GEPR-3s and WARs. Whether audiogenic kindling alters the patterns of brain structure in the GEPR-3s and WARs remain to be evaluated. Funding: Financial Support NGC:  FAPESP, CAPESP-PROEX, CNPq, PN: AA020073, PAF: KL2TR001432, R01NS097762