G. Weiss et al., IRON REGULATES NITRIC-OXIDE SYNTHASE ACTIVITY BY CONTROLLING NUCLEAR TRANSCRIPTION, The Journal of experimental medicine, 180(3), 1994, pp. 969-976
Recently, it was reported that nitric oxide (NO) directly controls int
racellular iron metabolism by activating iron regulatory protein (IRP)
, a cytoplasmic protein that regulates ferritin translation. To determ
ine whether intracellular iron levels themselves affect NO synthase (N
OS), we studied the effect of iron on cytokine-inducible NOS activity
and mRNA expression in the murine macrophage cell line J774A.1. We sho
w here that NOS activity is decreased by about 50% in homogenates obta
ined from cells treated with interferon gamma plus lipopolysaccharide
(IFN-gamma/LPS) in the presence of 50 mu M ferric iron [Fe(3(+))] as c
ompared with extracts from cells treated with IFN-gamma/LPS alone. Con
versely, addition of the iron chelator desferrioxamine (100 mu M) at t
he time of stimulation with IFN-gamma/LPS increases NOS activity up to
2.5-fold in J774 cells. These effects of changing the cellular iron s
tate cannot be attributed to a general alteration of the IFN-gamma/LPS
signal, since IFN-gamma/LPS-mediated major histocompatibility complex
class II antigen expression is unaffected. Furthermore, neither was t
he intracellular availability of the NOS cofactor tetrahydrobiopterin
altered by treatment with Fe(3(+)) or desferrioxamine, nor do these co
mpounds interfere with the activity of the hemoprotein NOS in vitro. W
e demonstrate that the mRNA levels for NOS are profoundly increased by
treatment with desferrioxamine and reduced by Fe(3(+)). The half-life
of NOS mRNA appeared not to be significantly altered by administratio
n of ferric ion, and NOS mRNA stability was only slightly prolonged by
desferrioxamine treatment. Nuclear run-off experiments demonstrate th
at nuclear transcription of cytokine-inducible NOS mRNA is strongly in
creased by desferrioxamine whereas it is decreased by Fe(3 C). Thus, t
his transcriptional response appears to account quantitatively for the
changes in enzyme activity. Our results suggest the existence of a re
gulatory loop between iron metabolism and the NO/NOS pathway.