Abstracts

Rationale: Tuberous Sclerosis Complex (TSC) is a developmental disorder associated with a high incidence of epilepsy. TSC is caused by loss of function mutations in either the TSC1 or TSC2 genes whose protein products form a complex that is a key negative regulator of the mTOR pathway. Mouse models of TSC have provided insights into possible disease mechanisms but do not recapitulate all aspects of the disease, especially the formation of "tubers", which often become seizure foci in patients. To address this, we have established a stem cell-derived human neuronal model of TSC that recapitulates the core cellular features of the disorder. This model provides a novel platform for investigating epileptogenesis in TSC and related disorders. Methods: To generate a human neuronal model for TSC we have engineered, by genome editing with CRISPR/Cas9, loss of function mutations in TSC1 and TSC2 in human embryonic stem cells (hESCs). We have differentiated these stem cells into neurons using dual-SMAD inhibition to create two-dimensional cultures of excitatory neurons and three-dimensional "cortical organoids". We used western blotting and immunofluorescence to determine the effect of TSC1 and TSC2 loss on stem-cell derived human neural precursors and neurons. Results: Differentiating hESCs with constitutive loss of TSC1 or 2 into neurons revealed a distinct morphology, including increased cell body size and altered dendritic arborization. We also observed differential protein expression, particularly in downstream targets of mTOR that control cellular metabolism and protein translation. Additionally, we observed that mosaic deletion of TSC2 during cortical organoid differentiation led to the appearance of tuber-like cells. These cells have high mTOR activity and markedly different morphology, resembling dysmorphic cells seen in resected patient tubers. Conclusions: Our approach provides a novel, genetically controlled platform from which to elucidate the cellular and molecular basis of TSC in a human neuronal context. Due to the high similarity between cellular phenotypes observed in our model and those reported in patient brain tissue, we expect that this will be a valid system for future functional characterization and therapeutic testing. Funding: This abstract was funded by a Predoctoral Research Fellowship from the American Epilepsy Society to JDB. It was additionally funded by a Brain Research Foundation Seed Grant (BRFSG-2014-02), an NINDS R01 (R01-NS097823) and a Hellman Family Faculty Fund grant to HSB.