Functional characterization of a human epilepsy-associated gene network reveals metabolic regulation as a critical factor underlying seizure susceptibilities

Epilepsy is a mechanistically complex, incompletely understood neurological disorder. To uncover novel converging mechanisms in epilepsy, we used Drosophila whole-brain single-cell RNA sequencing to refine and characterize a previously proposed human epilepsy-associated gene co-expression network (GCN). We identified a conserved co-expressed module of 26 genes, which comprises fly orthologs of 13 epilepsy-associated genes and integrates synaptic and metabolic functions. Over one-third of the Drosophila pan-neuronal knockdown models targeting this module exhibited altered seizure-like behaviors in response to mechanical or heat stress. These knockdown models recapitulated seizures associated with four epilepsy-associated genes and identified two novel epilepsy candidate genes and three genes for which knockdown conferred seizure protection. Most knockdown models with altered seizure susceptibility showed changes in metabolic rate and levels of phosphorylated adenosine monophosphate-activated protein kinase (AMPK), a key regulator of cellular energy homeostasis. Enhancing AMPK activity increased seizure resistance in a dose-dependent manner. Our findings show that Drosophila single-cell expression data and behavior can aid functional validation of human GCNs and highlight a role for metabolism in modifying seizure susceptibility.