Abstracts

Rationale: Idiopathic epilepsies (IEs) are common genetic disorders in both human and dogs. Most of the known human epilepsy genes have been found in rare monogenic Mendelian forms of epilepsy syndromes and the genetic background of common IEs remains largely unknown. Epilepsy is the most common neurological disorder in dogs providing natural models for human epilepsies. The first five genes discovered in dogs turned out to be the orthologs of the corresponding human progressive myoclonus epilepsies. IE genes have not yet been identified in dogs but the unique breed structure is anticipated to aid in gene mapping and lead to the discovery of epilepsy-causing mutations in different genes whose orthologs will likely contribute to the development of human IEs. Methods: We have established large sample cohorts and pedigrees for 20 different dog breeds suffering from idiopathic epilepsies. Clinical phenotyping was based on epilepsy questionnaires and neurological studies with EEG and MRI for selected dogs. Genetic analyses were performed for a varying number of cases and controls in each breed by genome-wide association mapping using Affymetrix’s or Illumina’s canine SNP chips arrays. Genotyping data were analyzed with PLINK software and positive associations were followed by fine-mapping studies and candidate gene sequencing. Results: Clinical analyses identify different types of focal idiopathic epilepsies in most of the studied breeds. We map and confirm several novel genome-wide significant associations to different chromosomes in several dog breeds including Lagotto Romagnolo, Belgian Shepherds, Greater Swiss Mountain Dogs, Finnish Spitzs and Norwich Terriers. Only one locus contains a previously mapped epilepsy gene, while the remainder notably contain no genes previously associated with epilepsy in any species. In some breeds, the preliminary associations are promising although did not yet reach genome-wide significance. In most breeds the identified loci are breed-specific but we also found some loci that are shared between breeds. Our data suggests that majority of canine IEs are oligogenic with multiple affected chromosomes. Finally, we identify a novel epilepsy-causing mutation on CFA3 in Lagotto Romagnolos with a recessive form of focal juvenile epilepsy with a benign outcome. The identified gene is a novel non-ion channel gene and is currently being screened in human epilepsy patients. Conclusions: Identification of a novel mutation and several other IE loci indicates the power of dog breeds as a novel approach to facilitate gene mapping in IEs. Most of the identified loci have not been associated with epilepsy, contain ion channel genes and are therefore likely to reveal several new previously unknown epilepsy genes that can be followed in human IE cohorts. These studies will likely establish several new canine models of human IEs and eventually improve the understanding and treatment of the conditions in both species.