Abstracts

Rationale: Recent studies have identified brain somatic variants as a cause of focal epilepsy. These findings were based on analyses of tissue obtained through epilepsy surgery, a procedure not available for most patients. The use of DNA derived from brain tissue adherent to SEEG depth electrodes has been proposed as an alternative but is hampered by the low cell quality and contamination by non-brain cells. We aimed to: (1) improve the SEEG-harvesting technique by purifying neuronal nuclei; (2) establish stringent quality control measures; (3) and implement an optimized bioinformatic workflow to reliably detect somatic variants.



Methods: SEEG electrodes were collected upon extraction from 41 brain regions across 17 patients undergoing SEEG in our center. Depth electrodes were pooled based on brain region and seizure onset zone. Nuclei were isolated separately from depth electrodes in the affected brain region (seizure onset zone) and the unaffected brain region (area least involved on SEEG and distant from possible MRI lesion) of epilepsy patients. This was followed by isolation of neuronal nuclei using Fluorescence-Activated Nuclei Sorting and DNA amplification using Primary Template Amplification. High-quality amplified DNA samples from affected brain regions, patient-matched unaffected brain regions and genomic DNA were subjected to whole exome sequencing (WES). Pre-sequencing Short Tandem Repeat (STR) analysis and post-sequencing allelic imbalance assessment were used to evaluate sample integrity. A bioinformatic workflow was developed to reduce false positives and to accurately detect somatic variants in the affected brain region.




Results: Based on DNA yield and STR analysis, 14 SEEG-derived neuronal DNA samples (7 affected and 7 unaffected) across 7 patients were chosen for WES. Allelic imbalance analysis revealed that of the 7 patient samples, 3 were of moderate and 4 of high quality, aligning with STR analysis results. Using our bioinformatic workflow, we prioritized 9 and 23 variants on average in the high and moderate quality samples, respectively. From these prioritized variants, we chose 4 candidate variants (MTOR: ENST00000361445:c.1465C >A, p.Leu489Met (Klein et al. Epilepsia 2024), CSDE1: ENST00000438362:c.1444G >T, p.Ala482Ser, KLLN: ENST00000445946:c.3G >A, p.Met1? and NLE1: ENST00000442241:c.1403G >C, p.Gly468Ala) across 4 patients (2 of moderate and high quality each) based on pathogenicity scores and association with phenotype. All 4 variants were validated using Digital Droplet Polymerase Chain Reaction.

Conclusions: Our results confirm the efficacy of our improved workflow to reliably detect somatic variants from neuronal DNA derived from depth electrodes. Our approach streamlines somatic variant discovery in focal epilepsies which will help elucidate the mechanisms of focal epilepsies, potentially leading to more effective treatments.


Funding: Canada Brain Research Fund (Government of Canada/Health Canada, Brain Canada, Azrieli Foundation), Epilepsy Canada, Department of Clinical Neurosciences, University of Calgary, Hotchkiss Brain Institute, Alberta Children’s Hospital Research Institute and Foundation, CIHR bridge grant, Canada Foundation for Innovation Grant.