Translatable EEG Measures in Mouse Models of Tuberous Sclerosis Complex (TSC)

Background: Tuberous sclerosis complex (TSC) is a rare genetic disorder associated with autism spectrum disorder (ASD) and epilepsy. Dysregulation of the mTOR pathway in TSC results in a loss of white matter integrity evident in imaging of patients, which is thought to be caused by alterations in axon outgrowth and synapse formation seen in animal and cellular models of Tsc1/Tsc2 mutation. Critical to harnessing the translational potential of animal models for drug discovery is the development of biomarkers that can be measured in both people and animals with high fidelity. The consequences of white matter disruption can be explored at the systems level in both animal models and patients through the measurement of neural activity via electroencephalography (EEG) to identify signatures of impairment associated with specific behavioral phenotypes and in response to pharmacological intervention.

Objectives: To identify translationally conserved alterations in neural activity associated with TSC related mutations, we characterized EEG based measures in transgenic mouse models of TSC that could be analogously measured in the patient population.

Methods: Utilizing mouse lines expressing a hypomorphic Tsc2 allele or a restricted Tsc1 mutation associated with selective social deficits, chronic in vivoEEG was recorded under baseline conditions and after either acute or chronic treatment with an mTOR inhibitor. Spectral features and seizure activity were characterized longitudinally over development and auditory evoked potentials elicited from auditory gating and mismatch negativity paradigms were extracted in adulthood.

Results: Mice with reduced expression of Tsc2 have lower oscillatory power in low frequency bands (theta) but an increase in power in the high frequency bands (beta and gamma), which evolves over development. The changes in spectral power are associated with a progressive seizure phenotype that ultimately results in the premature death of the animals. Paradoxically, chronic treatment with an mTOR inhibitor reduces the seizure burden of the animals but exacerbates the spectral phenotype. Mutant animals also have alterations in the evoked response to auditory tones, relative to control animals, characterized by enhanced early components and diminished late components of the response waveform that are analogous to the responses seen in the patient population. Changes in spectral power and auditory evoked potentials are induced with chronic but not acute rapamycin treatment that parallel functional rescue.

Conclusions: Features of resting state EEG and auditory evoked potentials in TSC mouse models both recapitulate early findings in patients with TSC and are responsive to treatment with mTOR inhibitors. Upon validation, these features could be used to drive biomarker selection for ongoing clinical trials with mTOR inhibitors in TSC. More broadly, the conservation of neurophysiological phenotypes across animal models and patient populations in TSC supports the use of EEG as a translational modality for exploring neural deficits associated with connectivity and synaptopathy.