Negative modulation of nitric oxide synthase by nitric oxide and nitroso compounds.

These observations clearly indicate that NO inhibits NOS activity and that nNOS and eNOS are more sensitive than iNOS to the inhibitory action of NO. Not only exogenously added NO but also enzymatically generated NO inhibits the activity of nNOS and eNOS. The mechanism by which NO inhibits NOS appears to involve the heme iron prosthetic group of NOS. Moreover, the oxidation state of the heme iron is critical in determining the magnitude of inhibition of NOS by NO. Conditions that favor the higher oxidation state of FeIII markedly increase the inhibitory action of NO, whereas conditions that favor the lower oxidation state of FeII markedly decrease the inhibitory action of NO. One of the cofactor roles of tetrahydrobiopterin may be to reduce the negative-feedback effect of NO on NOS by favoring the formation of the ferrous heme state in NOS. The inhibitory influence of NO on eNOS, albeit indirectly, was also observed in vascular endothelial cells, arterial rings, and in vivo in the perfused rabbit hindquarters vascular bed. Excess NO in the form of NO donor compounds inhibited the endothelium-dependent formation of EDRF/NO in response to endothelium-dependent vasorelaxants such as acetylcholine and bradykinin without influencing the relaxant effect of NO itself. These studies are consistent with the view that enzymatically generated NO may play an important negative-feedback regulatory role on eNOS, and therefore on vascular endothelial cell function. Several biological implications of a negative-feedback modulatory effect by NO on constitutive isoforms of NOS are evident. In nonadrenergic-noncholinergic transmission, in which NO is believed to be the principal inhibitory neurotransmitter (Sanders and Ward, 1992; Rand, 1992; Rajfer et al., 1992), NO may regulate its own synthesis, and therefore the neurotransmission process. Excess NO production may be undesirable because of the potential of NO or a reaction product of NO to elicit cytotoxic effects. Many extraneuronal factors could also contribute to decreasing the potentially cytotoxic actions of NO. For example, reduced hemoproteins such as hemoglobin, myoglobin, and/or their oxygen adducts could inactivate NO, as could superoxide anion generated in the vicinity of NO. In vascular endothelial cells either enzymatically generated NO or the presence of exogenously added NO in the form of nitrovasodilator drugs could diminish the vasodilator responses to endothelium-dependent relaxants and flow or shear stress. Although iNOS is less sensitive than either eNOS or nNOS to inhibition by NO, the generation of relatively large quantities of NO by iNOS within the confines of a cell may lead to a negative-feedback effect. The concomitant generation of superoxide anion by the same or adjacent cells could result in a diminished negative-feedback effect because of the rapid reaction between NO and superoxide anion to form peroxynitrite. Thus, NO production would increase and there would be increased peroxynitrite formation as well, which would result in enhanced cytotoxicity, provided that peroxynitrite is a cytotoxic species. Alternatively, iNOS may be conveniently insensitive to NO in order to allow for the generation of large quantities of NO for the purpose of producing cytotoxic effects.