Site and mechanism of uncoupling of nitric-oxide synthase: Uncoupling by monomerization and other misconceptions.

Nitric oxide synthase (NOS) catalyzes the transformation of l-arginine, molecular oxygen (O), and NADPH-derived electrons to nitric oxide (NO) and l-citrulline. Under some conditions, however, NOS catalyzes the reduction of O to superoxide (O) instead, a phenomenon that is generally referred to as uncoupling. In principle, both the heme in the oxygenase domain and the flavins in the reductase domain could catalyze O formation. In the former case the oxyferrous (Fe(II)O) complex that is formed as an intermediate during catalysis would dissociate to heme and O; in the latter case the reduced flavins would reduce O to O. The NOS cofactor tetrahydrobiopterin (BH4) is indispensable for coupled catalysis. In the case of uncoupling at the heme this is explained by the essential role of BH4 as an electron donor to the oxyferrous complex; in the case of uncoupling at the flavins it is assumed that the absence of BH4 results in NOS monomerization, with the monomers incapable to sustain NO synthesis but still able to support uncoupled catalysis. In spite of little supporting evidence, uncoupling at the reductase after NOS monomerization appears to be the predominant hypothesis at present. To set the record straight we extended prior studies by determining under which conditions uncoupling of the neuronal and endothelial isoforms (nNOS and eNOS) occurred and if a correlation exists between uncoupling and the monomer/dimer equilibrium. We determined the rates of coupled/uncoupled catalysis by measuring NADPH oxidation spectrophotometrically at 340 nm and citrulline synthesis as the formation of [H]-citrulline from [H]-Arg. The monomer/dimer equilibrium was determined by FPLC and, for comparison, by low-temperature polyacrylamide gel electrophoresis. Uncoupling occurred in the absence of Arg and/or BH4, but not in the absence of Ca or calmodulin (CaM). Since omission of Ca/CaM will completely block heme reduction while still allowing substantial FMN reduction, this argues against uncoupling by the reductase domain. In the presence of heme-directed NOS inhibitors uncoupling occurred to the extent that these compound allowed heme reduction, again arguing in favor of uncoupling at the heme. The monomer/dimer equilibrium showed no correlation with uncoupling. We conclude that uncoupling by BH4 deficiency takes place exclusively at the heme, with virtually no contribution from the flavins and no role for NOS monomerization.