Abstracts

Rationale: Tuberous Sclerosis Complex (TSC) is caused by variants in TSC1 and TSC2, leading to mTOR pathway hyperactivation. Up to 90% of patients develop epilepsy, with TSC2 variants linked to more severe disease. Phenotypic heterogeneity exists among TSC2 variant carriers likely due to second genetic hits, mosaicism, and the nature of the primary mutation. The functional consequences of primary mutations are largely unexplored. Here, we assess the effects of different TSC2 variants on neuronal morphology and mTOR signaling.

Methods: We used transgenic mice with a floxed Tsc2 gene and a Cre-dependent fluorophore (Tsc2^flox/flox, tdTomato^+/-) combined with a lentiviral strategy to delete the murine Tsc2 allele and express human FLAG-tagged TSC2 (hTSC2) variants. We selected 3 TSC2 variants (hTSC2^var): p.Y1678* (nonsense, severe epilepsy, GAP domain), p.R905W (missense, mild epilepsy, domain function unknown), and p.R905Q (missense, severe epilepsy, domain function unknown). Variants were tested in vitro in cultured cortical neurons and in vivo in mice. For in vitro experiments, neurons were transduced with Cre-only virus (Tsc2KO), Cre-hTSC2^wt, or Cre-hTSC2^var at 7 days in culture. After 14 days, we used qRT-PCR, western blot, and immunocytochemistry to confirm deletion of mTsc2, confirm transgene expression and assess the effects on morphology and mTOR activity. For in vivo experiments, lentiviral vectors were injected into the cortex of neonatal mice. After 8 weeks, brains were analyzed for Tsc2 KO, transgene expression, mTOR activation, and neuronal morphology.


Results: QRT-PCR analyses using Cre-, hTSC2- and FLAG-specific primers verified transgene expression and mTsc2 gene knockout as expected in all groups. Western blot analyses confirmed murine Tsc2 deletion and re-expression of hTSC2. Fluorescence immunocytochemistry showed increased mTOR target S6 phosphorylation and soma size in Tsc2KO neurons (n=160-260 cells, p< 0.001). Cre-hTSC2^WT normalized these to wild-type levels (p< 0.001). All missense variants increased S6 phosphorylation and soma size, though less than Tsc2KO (Tukey’s posthoc < 0.0001), suggesting partial loss of function. Preliminary results suggest that soma size in vitro correlated with patient’s epilepsy severity with both severe variants (p.Y1678* and p.R905W) having larger somas than the p.R905Q variant associated with a milder phenotype (Tukey’s posthoc < 0.05). Likewise, S6 hyperphosphorylation may be more pronounced in the truncated variant p.Y1678* than the two missense variants p.R905W and p.R805Q (Tukey’s posthoc < 0.0001). In vivo experiments are ongoing. No early mortality or morbidity associated with mTsc2 KO or hTSC2 variants has been observed.

Conclusions: We established a platform to characterize human TSC2 variants in mice, suggesting that the studied missense variants lead to partial rather than complete loss of function. Disease severity may correlate with the degree of mTOR hyperactivity and cellular dysmorphology. Future in vivo experiments will assess each variant's effects on epilepsy. By testing additional variants, we aim to establish genotype-phenotype relationships to predict epilepsy severity.

Funding: Cincinnati Children's Hospital Medical Center