Authors :
Devin Cortes, BS – University of Pittsburgh
Noah Coulson, BS – University of Pittsburgh
Dalton West, BS – University of Pittsburgh
Thomas Becker-Szurszewski, BS – University of Pittsburgh
Sean Hartwick, BS – University of Pittsburgh
Sivakama Bharathi, PhD – University of Pittsburgh
Kevin Kelly, MD – Allegheny Health Network
Anthony Christodoulou, PhD – University of California Los Angeles UCLA
Eric Goetzman, PhD – University of Pittsburgh
Presenting Author: Yijen Wu, PhD – University of Pittsburgh
Rationale:
Temporal lobe epilepsy (TLE), the most frequent form of acquired epilepsy, is typically initiated by an insult, such as status epilepticus (SE), followed by a latent period wherein molecular and cellular remodeling occurs leading to chronic epilepsy. Not every patient who experiences episodic SE will progress to TLE. There is an unmet clinical need to prevent irreversible refractory epilepsy by understanding, detecting, and reducing epileptogenesis. Mitochondrial dysfunction is increasingly recognized as an inciting factor for TLE, not only acutely after SE, but also contributing to epileptogenesis for refractory TLE. We have successfully developed a novel 4D Oxy-wavelet MRI capable of detecting mitochondrial dysfunction in a spatially specific manner. 4D Oxy-wavelet MRI is spatially resolved throughout the whole brain, permitting network analysis. This is a critical capability, as epilepsy is well recognized as a network disorder, wherein seizure generation, spread, and cognitive impairment can arise from the dysfunction of large-scale networks. Methods:
In a rat TLE model, we conducted 4D Oxy-wavelet MRI at acute after the initial kainic acid (KA) induced status epilepticus (SE) injury to predict animals that would develop chronic epilepsy after 3 months. 12 males and 12 female Sprague-Dawley rats were subjected to multiple low-doses of kainic acid (KA 5mg/kg then 2.5mg/kg) to elicit Racine scale 3/4/5 SE for 90 minutes. Rats were subjected to 4D Oxy-wavelet MRI before (baseline) and on 3, 7, 10, 14, 21 days after KA induction. Rats were then single housed for continuous video recording for spontaneous seizures for 2 months. Neuronal network analysis was performed to delineate functional networks related to epileptogenesis.
Results:
Out of the 24 rats, 7 (4 males and 3 females) developed chronic TLE (Fig.1A,C), and the other 17 rats did not (Fig.1B,D). On day 10 after KA induction, TLE animals (Fig.1AC bottom row) showed mostly bad Oxy-wavelet index mapping, but the group-averaged Oxy-wavelet index maps of no-TLE animals (Fig.1BD bottom row) showed marked improvement. Hierarchical clustering analysis (Fig.2A) with 3 eigenvectors for all 24 animals covering all acute time points (days 3, 7, 10, 14, 21) could parcellate animals into 2 clusters: the “more injured” (Fig.2A red) and the “less injured” (Fig.2A blue) clusters. 6 out of the 7 TLE rats are in the “more injured” cluster (Fig.2A red), whereas 11 out of 12 non-seizing rats are in the “less injured” cluster (Fig.2A blue). Adjusted p value maps on Day 10 (Fig.2B) for the “more injured” cluster revealed brain regions with significant mitochondrial injury, including hippocampus, amygdala, somatosensory cortex, piriform cortex, and thalamus, consistent with temporal lobe epilepsy and hippocampal-thalamic projection.
Conclusions:
4D Oxy-wavelet MRI at acute time window can differentiate epilepsy-prone animals that would develop chronic epilepsy 3 months later, providing a non-invasive biomarker for epileptogenesis.
Funding:
NS121706 (NIH) and W81XWH-22-1-0221(DoD)