Abstracts

Rationale: Growing evidence demonstrates a role for miRNAs in the pathomechanism of epilepsy. We hypothesize that variants in brain-expressed miRNAs can alter the expression of their target genes, giving rise to the severe phenotype of the epileptic encephalopathies (EE). Methods: We screened 508 brain-expressed miRNA genes in DNA derived from EE patients and their healthy parents using 'Multiplex Amplification of Specific Targets for Resequencing' (MASTR) assays (Multiplicom NV). Furthermore, we annotated unsolved whole exome sequencing data of patient-parent trios with a miRNA annotation database and filtered on de novo variants present in primary miRNAs. Results: So far, we identified one de novo variant and one mosaic de novo variant in two distinct miRNAs genes. Both variants are present in the mature miRNA and predicted to disrupt the secondary structure of the miRNA. In order to evaluate the impact of the variants on the mature miRNA levels, we generated SH-SY5Y cells overexpressing wild type or variant miRNA and performed both mRNA sequencing and small RNA sequencing. For the de novo miRNA variant, the small RNA sequencing data revealed arm switching of the predominant mature miRNA and the formation of 5' isomirs. Isomirs are variations in mature miRNAs with respect to their canonical miRNA as stated in miRBase. 5' isomirs result in seed shifting, presumably leading to altered target recognition and regulation. In addition, pathway analysis of the differentially expressed genes in the mRNA-seq data showed an enrichment of pathways involved in neuronal development and epilepsy. Analysis of the mosaic de novo variant showed that this variant miRNA was weakly expressed compared to wild type expression. Pathway analysis of the mRNA sequencing data revealed no enrichment of relevant neuronal pathways. We validated the mRNA sequencing results for the de novo variant using QuantSeq (Lexogen), including RNA from 28 additional cell pellets (14 expressing wild type miRNA and 14 expressing variant miRNA). The significant differentially expressed genes showed a correlation of 0.80 between the mRNA sequencing and QuantSeq data. Further analysis is ongoing in order to acquire more evidence linking the variant in the miRNAs with the EE phenotype. Conclusions: These findings show that brain-expressed miRNAs are interesting candidate genes for EE and are worthwhile to investigate. Funding: This work has been supported by grants from the Institute for Science and Technology (IWT)-Flanders.