The previous biochemical evidence had suggested that glutaric aciduria type II (GA II) is due to deficient dehydrogenation of multiple short-chain acyl coenzyme A's (CoA's), bu the precise biochemical mechanism underlying this disease was unknown. We investigated substrate oxidation and in vitro activities of isovaleryl CoA- and butyryl CoA dehydrogenases as well as that of electron-transferring flavoprotein (ETF) in cultured skin fibroblasts from a patient with GA II. GA II cells have a markedly decreased ability to oxidize [1-14C]butyrate, [2-14C]lysine, and [2,14C]leucine (3, 9, and 9% of control values, respectively). Mitochondrial isovaleryl CoA- and butyryl CoA dehydrogenase activities in GA II cells were determined using a tritium release assay with [2,3-3H] acyl-CoA's as substrate. When an artificial electron acceptor, phenazine methosulfate (PMS) was not added in the assay media, these activities were 108 and 113% of controls, respectively. This represents the normal abilities of the dehydrogenases in GA II cells to bind the substrate and to catalyze tritium exchange between the bound substrate and solvent. When PMS was added to the assay mixture, these activities were 88 and 70% of control values, respectively, indicating that these enzymes can both dehydrogenate their substrates normally and then transfer electrons to an acceptor (PMS). ETF activity in mitochondrial sonic supernatants from GA II cells, as assessed by a newly devised method, was 159% of control values. These observations suggest that the acyl CoA dehydrogenases themselves and ETF are not defective in GA II. Therefore, the deficiency of another common gene product necessary for the function of all the affected acyl CoA dehydrogenases must be sought to explain the etiology of GA II.