Pyridox(am)ine 5’-phosphate oxidase deficiency induces seizures inDrosophila melanogaster

Abstract

Pyridox(am)ine 5'-phosphate oxidase (PNPO) is a rate-limiting enzyme in converting dietary vitamin B6 (VB6) to pyridoxal 5'-phosphate (PLP), the biologically active form of VB6, and involved in the synthesis of neurotransmitters including GABA, dopamine, and serotonin. In humans, PNPO mutations have been increasingly identified in neonatal epileptic encephalopathy and more recently also in early-onset epilepsy. Till now, little is known about the neurobiological mechanisms underlying PNPO-deficiency-induced seizures due to the lack of animal models. Previously we identified a c.95 C > A missense mutation in sgll -the Drosophila homolog of human PNPO (hPNPO) and found mutant (sgll 95 ) flies exhibiting a lethal phenotype on a diet devoid of VB6.Here we report the establishment of both sgll 95 and ubiquitous sgll knockdown (KD) flies as valid animal models of PNPO-deficiency-induced epilepsy. Both sgll 95 and sgll KD flies exhibit spontaneous seizures before they die. Electrophysiological recordings reveal that seizures caused by PNPO deficiency have characteristics similar to that in flies treated with GABA antagonist picrotoxin. Both seizures and lethality are associated with low PLP levels and can be rescued by ubiquitous expression of wild-type sgll or hPNPO , suggesting the functional conservation of the PNPO enzyme between humans and flies. Results from cell type-specific sgll KD further demonstrate that PNPO in the brain is necessary for seizure prevention and survival. Our establishment of the first animal model of PNPO deficiency will lead to better understanding of VB6 biology, the PNPO gene and its mutations discovered in patients, and can be a cost-effective system to test therapeutic strategies.

Glial expression ofDrosophila UBE3Acauses spontaneous seizures modulated by 5-HT signaling

“…Automated video tracking approaches have been employed to characterize walking behavior in several hyperexcitable mutants (Chi et al, 2019;Iyengar et al, 2012;Kaas et al, 2016;Stone et al, 2013). Here, we built upon this work to identify seizure-related behavioral events based on locomotion tracking using a custom machine-learning approach.…”