Genotype-informed single cell transcriptomic profiling of mosaic brain tissue

Somatic variants that arise during embryonic corticogenesis and result in brain mosaicism are increasingly recognized as significant contributors to the genetic risk of neurodevelopmental and neuropsychiatric conditions. The discovery of disease-causing somatic variants in the brain has not only contributed to the identification of novel genes and refined our understanding of the underlying genetic architecture of a range of neuropsychiatric diseases, but they also provide powerful research tools that can be used to understand how cell-type-specific changes contribute to disease pathophysiology. In this study, we seek to use single-cell genomic approaches in mosaic human brain tissue to establish which cell types harbor the disease-causing somatic variants and to determine cell-type-specific transcriptomic changes associated with the variant. While single-cell genomics approaches have been used to inform many novel aspects of pathophysiology across a range of diseases, no studies to our knowledge consider genotype and use deep single-cell RNA sequencing to study variant- associated transcriptomic changes in specific cell types from mosaic human brain tissue. Here, we propose to perform a proof-of-concept study to establish the extent to which genotype-informed comprehensive single-cell RNA sequencing of mosaic human brain tissue is reproducible and can inform novel aspects of disease pathophysiology. We will accomplish this goal using hemimegalencephaly, a severe brain malformation causing overgrowth of one cerebral hemisphere that is caused by a single highly penetrant, brain tissue-specific somatic variant in either AKT3, MTOR, or PIK3CA, as our prototype. We will isolate both DNA and RNA from single nuclei from therapeutically resected mosaic brain tissue of each individual, genotype the DNA to classify each isolated cell as variant-positive or variant-negative, and then perform RNA sequencing to determine cell-type-specific variant burden and define the transcriptomic changes associated with the variant in specific cell types. We will analyze three individuals with genetically diagnosed hemimegalencephaly caused by the same pathogenic somatic variant for each of the three known disease genes to allow for comparisons across individuals and genes. Comparisons of variant-positive to variant-negative cells in hemimegalencephaly cases, along with variant-negative cells from age-matched neurotypical control brain tissue, within like cell types, will allow us to identify cell type-specific cell-autonomous and non-cell-autonomous transcriptomic changes associated with the variant. We hypothesize that these studies will reveal reproducible convergent and divergent cell-type-specific disease mechanisms across individuals with the same pathogenic variants and across genes implicated in the same brain malformation. If successful, we will establish a powerful research approach that leverages somatic mosaicism to inform cell types involved in disease, identify pathophysiologically relevant specific cell-type specific transcriptomic changes involved in neurological and neuropsychiatric diseases, and illuminate novel therapeutic strategies in hemimegalencephaly and related conditions.