As the primary inhibitory neurotransmitter in the central nervous system, γ-aminobutyric acid (GABA) plays a critical role in controlling neural activity. Dysregulation of GABA synthesis and release or disruptions in the excitation and inhibition balance of the brain have devastating pathological consequences, such as epilepsy. Epilepsy is a highly heterogeneous neurological condition which can stem from purely genetic, gene-environment interaction, or non-genetic etiologies. Since discovery of the first epilepsy-associated gene in 1995, nearly 1000 contributing loci and mutations have been found. However, how individual mutations may lead to excitation and inhibition imbalance and seizures remains poorly understood. One group of drug-resistant epilepsies occurs in human patients with mutations in the pyridox(am)ine-5’-phosphate oxidase (PNPO) gene. PNPO metabolizes dietary inactive forms of vitamin B6 into the active pyridoxal-5’-phosphate (PLP), which is a critical cofactor for synthesis of GABA. We generated two novel genetic knock-in mouse models containing PNPO point mutations identified in human epilepsy patients: D33V and R116Q. Homozygous D33V mutants require supplemental PLP feeding for survival and exhibit spontaneous seizures around P15. Meanwhile, homozygous R116Q and heterozygous D33V do not have spontaneous seizures but exhibit decreased latency to chemically-induced seizures, increased hyperactivity, and impaired spatial learning and memory. Using electroencephalography (EEG) and virally-expressed fluorescent GABA sensors, we found that heterozygous D33V mice exhibited increased theta rhythm power which could be rescued with PLP supplementation, increased ictal propagation speed across the cortex, and decreased GABA neurotransmitter release compared to wild-type controls. The data suggests that PNPO mutant mice recapitulate phenotypes exhibited in epilepsy patients, and that GABA deficiency in PNPO mutant mice contributes to a collapse of feedforward inhibitory control and increased seizure susceptibility. One of the largest contributors to new onset of seizures in humans is excessive alcohol use and withdrawal, and the variable behavioral and psychosocial consequences of alcohol consumption between individuals suggests a strong genetic component affecting alcohol response. In a second project, we examined the role of PNPO and PLP in alcohol use and behavioral response to alcohol consumption. In the central nervous system, one of the main targets of alcohol are GABAA receptors, where it acts as a positive allosteric modulator. Moreover, chronic alcohol consumption has been shown to cause deleterious effects on PLP content in humans. However, the intricate inter-relationship between alcohol use, PLP content, and GABAergic transmission has not yet been systematically explored. We previously generated and characterized knock-in fly models in which we replaced the fly PNPO gene with mutant human PNPO from epilepsy patients. Taking advantage of these fly models, we found that 1) alcohol consumption leads to PLP reduction; 2) PLP deficiency increases alcohol consumption; 3) PNPO mutations impair alcohol clearance; and 4) PNPO mutations have potentially lethal consequences which are worsened by alcohol consumption and rescued with PLP supplementation. Therefore, PNPO mutant flies exhibit an increase in alcohol consumption and decreased alcohol clearance, both of which lead to increased body alcohol and exacerbation of endogenous PLP deficiencies.