Molecular mechanisms in the epilepsies of infancy.

Over the past decade and a half, considerable advances have been made in the understanding of the molecular mechanisms underlying the idiopathic epilepsies. Technological advances and completion of the Human Genome Project have enabled continued progress. Much of this has impacted on families with epilepsies developing in infancy. Benign familial neonatal seizures (BFNS) is often caused by mutations in either of two potassium channel subunit genes, KCNQ2 and KCNQ3. Twenty-three of 36 families investigated (65%) were found to have mutations in one of these genes detectable by sequencing. Multiplex ligation-dependent probe amplification, which detects deletions and duplications affecting a specific gene, was applied to solve a further 17% of families. This revealed that deletions and duplications in KCNQ2 are a common mechanism for the pathogenesis of BFNS. The remaining unsolved BFNS families were analysed further to seek other mechanisms. A novel microduplication was identified in one family with BFNS and intellectual disability. This was characterised by comparative genome hybridisation (CGH) and fluorescence in-situ hybridisation and demonstrated the value of applying these technologies to familial as well as sporadic cases. A patient with neonatal seizures and long-QT syndrome (LQTS) was found to have a unique combination of changes in two genes associated with LQTS, supporting speculation that he had a “cardio-cerebral” channelopathy. Two “BFNS” families had mutations in SCN2A, the gene usually associated with benign familial neonatal-infantile seizures (BFNIS). BFNIS is distinguished from BFNS by a higher age of seizure onset distribution. The initial clinical misclassification highlights the phenotypic overlap between these two disorders. These families are now reclassified on molecular criteria as BFNIS families with an earlier than usual age of onset. This distinction is of clinical significance since unlike BFNIS with SCN2A mutations 15% of BFNS patients with KCNQ2 mutations have seizures later in life. The remaining three BFNS families are unsolved. For two families, genotyping of microsatellite markers linked to known BFNS loci showed that they could not have mutations at those loci. Linkage to these loci was excluded by recombination, demonstrating that at least one other gene associated with BFNS exists. The parental origin of de novo mutations in SCN1A was investigated. These mutations cause Dravet Syndrome (DS), a severe childhood epileptic encephalopathy. The mutations were found to originate on the paternal chromosome in approximately 75% of cases. The effect of parental age on mutagenesis in SCN1A was investigated and found not to be a contributing factor. This is the only epilepsy syndrome where sufficient de novo mutations have been identified for meaningful analysis of their parental origin. Finally, the causative gene for benign familial infantile seizures (BFIS) mapped to chromosome 16p11.2-q12.1 remains elusive. Array CGH revealed no pathogenic copy number changes. Sequence capture and next-generation sequencing of the genes in the linkage region did not detect a mutation in a coding region. Several unique, but nonpathogenic, variants were identified in BFIS families. This paves the way for the next steps aimed at detecting rarer molecular defects such as recurrent inversions or unstable repeats.Thesis (Ph.D.) -- University of Adelaide, School of Paediatrics and Reproductive Health, 201