Abstracts

Rationale: Massively parallel sequencing, or next-generation sequencing, has revealed the prominent role of de novo genetic mutations in causing the devastating epileptic encephalopathies (EEs). De novo mutations may arise post-zygotically in the developing fetus, resulting in a mosaic distribution of the mutation in the affected child. Alternatively, a post-zygotic mutation may arise in the parental germline, resulting in a constitutional mutation (where every cell carries the mutation) in the affected child. Mosaic parents may not have any noticeable signs of the disorder depending on the cell type and percentage of cells that carry the mutation, but germline mosaicism leads to a critical increase in recurrence risk in offspring. Often the clinical question of parental mosaicism only arises when the family has a second affected child, having previously been counseled with a low recurrence risk. While the importance of mosaicism is increasingly recognized, levels below 20% are difficult to identify with traditional Sanger sequencing and may escape detection. Methods: We used a highly sensitive, improved version of the molecular inversion probe technology called single molecule molecular inversion probes to investigate the frequency of somatic (and germline) mosaicism in parents of children with EEs associated with heterozygous de novo mutations. We screened 142 families where the affected child's EE was attributed to a substitution or small indel in one of 31 established epilepsy genes and reported as 'de novo' by either clinical or research analysis of parental DNA. We screened for the causative variant in DNA isolated from parental blood or saliva. Results: We identified 10 cases of low-level somatic mosaicism. The fraction of mutant alleles identified ranged from 0.1 ?" 11%, levels that would pass undetected by traditional Sanger sequencing methods. Mosaicism of the pathogenic variants was detected in six mothers and four fathers. In two families, two siblings had an EE with a mosaic CACNA1A mutation in one family and a mosaic DNM1 mutation in the other. Conclusions: The rate of parental mosaicism has been greatly underestimated in a sporadic patient with an EE and an apparent de novo mutation. This is because low levels ( < 11%) of mosaicism are found in 10% of cases and escape detection by traditional methodologies. Our findings highlight the critical importance of more sensitive tools for detecting low frequency mosaicism. Identifying of a mosaic mutation in parents is crucial for reproductive counseling and family planning. Funding: Funding for this study comes from the NIH National Institute of Neurological Disorders and Stroke 2R01NS069605 (H.C.M). C.T.M is supported by a postdoctoral fellowship provided by the Lennox-Gastaut Syndrome Foundation.