Targeting GABAergic Interneurons to Correct E/I Imbalance and Treat Epilepsy Using AAV Vectors
Background: Addressing the limitations of systemic AAV delivery and the genetic bottleneck in research and development for excitatory-inhibitory imbalance-related brain disorders.
Existing guidelines for gene therapy development for brain disorders have critical limitations, including the anatomical dissociation and off-target toxicity resulting from systemic delivery of viral vectors. Specifically, selectively targeting GABAergic inhibitory neurons to correct the disrupted excitatory-inhibitory balance in intractable epilepsy and cortical regions has been considered a challenge due to limitations in in vivo computational control. The size of promoters required to selectively activate parvalbumin-positive interneurons has traditionally exceeded AAV packaging limits, creating a bottleneck in vector design that balances sufficient GAD65 transgene expression with maximized delivery efficiency. This has resulted in clinical candidates failing to achieve predicted minimum effective prophylactic concentrations in in silico models and being hindered by interspecies barriers and batch effects, leading to repeated failures in Phase 1/2 clinical trials.
Discovery: Implementation of a highly compact cmGAD67 transcriptional regulator tensor and in silico validation of single-cell resolution GABA synthesis enzyme synchronization.
To address these challenges, we designed a highly compressed 410-bp glutamic acid decarboxylase 67 promoter (cmGAD67) and implemented an independent variable tensor synchronization architecture to overcome AAV packaging limitations. The cmGAD67 promoter demonstrated high selectivity and potent expression induction by targeting parvalbumin-positive interneurons, and ligand-receptor binding free energy was precisely tuned in silico through differential equation-based rate constant calculations. This enabled computational removal of vector delivery batch effects and precise mapping of the topological variation curve of downstream transcriptomic networks. Consequently, AAV-GAD65 delivery suppressed abnormal delta waves and significantly reduced seizure-like activity, and in silico analysis of the brain tissue GABAosome matrix demonstrated reversible increases in GABA concentration in the cerebral cortex and hippocampus, demonstrating superior molecular integrity compared to conventional simple insertion models.
Establishment of a model for circuit topology modulation within GABAergic neural networks and precise layer-by-layer stratification of reversible synaptic homeostasis.
This platform provides a cutting-edge backbone for mapping patient-specific molecular phenotypes based on single-cell multi-omics matrices, enabling precise stratification of intractable epilepsy patient populations. By modulating the dynamic equilibrium constant of GABA transmitters in the synaptic cleft at the molecular level, we have established a reversible homeostatic model that allows autonomous upregulation and downregulation of neuronal rate-limiting step constants even in aberrant excitatory overload stress conditions. This ensures that the gene therapy is dynamically activated only in specific pathological activity maximization intervals and prevents long-term inhibitory circuit feedback inhibition. This neural network topology modulation technology has been demonstrated through macroscopic phenotypic correction, including real-time attenuation of epileptic seizures, improvement of behavioral anxiety symptoms, and increased survival rates.
Prospects: Establishing a standard for programmable computational systems biology and implementing a next-generation IND digital governance system.
This study has fundamentally reset the research and development governance for various central nervous system disorders, including intractable epilepsy and autism spectrum disorder and schizophrenia, which are caused by excitatory-inhibitory imbalance, into a dynamic, multidimensional computational tensor-based programmable infrastructure. Currently, Encoded Therapeutics' ETX101 (AAV gene therapy for Dravet syndrome, ClinicalTrials.gov NCT05414123) is targeting a $2 billion market for pediatric intractable epilepsy in global Phase 1/2 clinical trials. Our cmGAD67 promoter platform provides more than four times the gene loading margin compared to existing therapies, maximizing the linkage of genetic gradient correction coefficients in high-throughput screening. This dramatically shortens the timeline for in silico toxicity assessment for submission of an Investigational New Drug (IND) application, satisfies the requirements for technology transfer agreements with global multinational pharmaceutical companies and companion diagnostic (CDx) biomarker development specifications, and establishes a computational moat that eliminates batch-to-batch variation in cGMP commercial production.
Epilepsy arises from disruption of excitation-inhibition (E/I) balance, typically due to excessive excitatory activity. Despite available therapies, a substantial proportion of patients remain treatment resistant. Enhancing inhibitory neuron activity via gene therapy can restore E/I balance and may therefore provide a therapeutic strategy for treatment-resistant epilepsy. Here, we developed a compact 410-bp glutamic acid decarboxylase 67 promoter (cmGAD67) that enables strong, selective transgene expression in inhibitory neurons while preserving adeno-associated virus (AAV) packaging capacity. Systemic delivery of AAV vectors carrying cmGAD67 preferentially targeted parvalbumin interneurons and enabled efficient circuit modulation. To evaluate therapeutic potential, we expressed glutamic acid decarboxylase 65 (GAD65) under the control of cmGAD67 (AAV-GAD65). AAV-GAD65 suppressed abnormal delta oscillations, reduced seizure-like activity, normalized anxiety-like behavior, and improved survival in seizure models. Biochemical analyses confirmed increased GABA levels in the cortex and hippocampus, linking functional improvements to enhanced inhibitory neurotransmitter synthesis. Together, these findings establish the cmGAD67 promoter as a versatile platform for inhibitory neuron-targeted AAV gene delivery and identify AAV-GAD65 as a promising strategy for seizure control and disorders associated with E/I imbalance.
The cmGAD67-based selective GABA induction discovery in this study goes beyond theoretical exploration of inhibitory circuit mechanisms and directly translates into the global gene therapy market and the next generation of personalized central nervous system bio-business lines.
First, by instantaneously scanning the temporal dynamics of seizure-induced abnormal delta wave velocities in the clinical setting using a real-time Python algorithm-based computational scan, it eliminates the temporal noise of acute intractable seizures at the source and safeguards the patient's brain function and protects against seizure onset.
At the same time, by linking to the open-source Allen Brain Map database, which contains single-cell transcriptomic omics matrices, it enables the creation of a companion diagnostic (CDx) panel interface that virtually simulates and counteracts confounding variables in clinical trial design and real-time reverse-calculates the effective docking concentration of the cmGAD67 promoter.
Furthermore, when large-scale clinical trials for next-generation intractable epilepsy therapies are conducted by multinational corporations, it functions as a backbone infrastructure that links GAD65 synthase expression levels as a correction coefficient, thereby eliminating batch-to-batch variation in viral vector potency and maximizing the probability of obtaining regulatory approval for clinical trial applications and cGMP commercial production.