Abstracts

Rationale: Juvenile myoclonic epilepsy (JME) is one of the most common epilepsy syndromes. While most patients respond well to antiepileptic medications, a minority of JME patients continue to have seizures despite treatment. Family history is a strong predictor of JME risk, yet few genetic variants have been indentified that clearly confer disease susceptibility. Deciphering the genetic underpinnings of this syndrome could greatly improve our understanding of the pathophysiologic processes governing this and other epilepsy disorders. Methods: In this study we sought to identify genetic variants that increase JME susceptibility using next-generation sequencing technology. Patients were selected for study if they had a diagnosis of JME and seizures that failed to be controlled with valproic acid, as determined by the treating physician. Agilent s SureSelect Human All Exon technology was used to capture the DNA regions that encode proteins from 50 refractory JME patients meeting the selection criteria. The captured fragments from each individual were then sequenced on one lane of an Illumina GAII sequencer. Sequenced fragments were aligned to the reference genome using BWA software. Single nucleotide variants (SNVs) and small insertion-deletions (indels) were called from the sequence data using SAMTools software. Annotation and statistical analyses were performed with SequenceVariantAnalyzer software. To ascertain the likelihood of variants causing JME, candidate variants observed in JME patients and infrequently observed in 100 exome or whole-genome sequenced controls not enriched for seizure phenotypes, were genotyped in 200 non-refractory JME patients, in more than 2,000 neuropsychiatrically-normal controls , and also in a subset of affected and unaffected family members of exome-sequenced JME patients. Results: Approximately 165,000 exons (~18,500 genes) were targeted with this technology and on average we successfully sequenced >93% of the bases in these exons with greater than 5-fold coverage. On average, 23,638 high-quality SNVs and 1,908 indels were identified in each sequenced JME exome. Furthermore, approximately 600 SNVs and 165 indels per JME patient are predicted to alter a protein-coding sequence and are not observed in any control sample. Targeted analysis of GABRA1 and EHFC1, Mendelian genes known to harbor genetic variants that cause JME, identified no rare, functional mutations in patients with JME. One rare heterozygous variant was observed in GABRA1 in one JME patient, however it is not predicted to have an effect on the coding sequence of the protein. Conclusions: We conclude that variants in Mendelian epilepsy genes cannot explain the majority of refractory JME cases studied here. Consistent with a model of locus heterogeneity in JME genetic susceptibility, we found no single rare functional variant that accounted for all or many of the studied cases. We continue to explore the possibility that rare variants increase the risk of refractory JME.