Ethanol is a widely used drug that is consumed in large amounts for pharmacologic effects. Elimination of ethanol alters metabolism in the liver and throughout an organism. Ethanol's effect on metabolites can influence the regulation of key pathways such as gluconeogenesis. We adopted a proton NMR-based metabolomic approach to study ethanol-induced metabolic changes in liver, blood, and brain tissue from rats exposed to either a single dose of ethanol or a chronic 4 day binge-ethanol protocol. Both acute and binge ethanol caused (i) decreased glucose, lactate, and alanine in liver and serum; (ii) increased acetate in liver and serum; and (iii) increased acetoacetate in serum. Binge-ethanol increased liver beta-hydroxybutyrate and decreased betaine. Pretreatment with the antioxidant butylated hydroxytoluene (BHT) increased betaine and reduced ethyl glucuronide (EtG) in livers of binge-ethanol animals, as compared to those not pretreated with BHT. We found no change in brain metabolites after a single dose of ethanol. Unsupervised principal component (PC) analysis of spectral data from liver and serum successfully discriminated treatment groups, based largely on the biochemical differences outlined above, confirming the results of manual analysis. To explain the observed lack of gluconeogenesis following ethanol treatment and to resolve apparently discordant results from previous studies, we propose a model in which decreased hepatic alanine removes inhibition on pyruvate kinase, thus permitting a futile cycle that diverts phosphoenolpyruvate away from gluconeogenesis. This is a new mechanism that biochemically elucidates the well-known, yet unexplained, "empty calorie" phenomenon of ethanol. Reduction of EtG by pretreatment with BHT suggests that BHT and perhaps other compounds may alter the pharmacokinetics of EtG so that EtG may not always be a sensitive marker for ethanol abuse.