This reaction is specific, that is, NarL-P is only desphosphorylated by NarX. The system discriminates between nitrate and nitrite. In the presence of nitrate both sensors phosphorylate NarL and NarP to the same extent. In the presence of nitrite, NarX phosphorylates NarP but acts primarily as a phosphatase for NarL-P [57, 58, 59]. In contrast, NarQ phosphorylates NarL and NarP in response to both nitrate and nitrite. Therefore, NarL mainly serves as a nitrate regulator and is only weakly phosphorylated in the presence of nitrite [59, 60]. NarL consists of the N-terminal receiver domain and a C-terminal DNA-binding domain. The crystal structures of the monomeric full-length protein [61, 62] and of the dimeric output domain bound to DNA [63, 64] have been reported. Based on the results of an EMSA and an in vivo expression assay, it was determined that the NarL OD (output domain) function has discrepant activities between NarL OD linkers. In the absence or presence of linker, the helix 6 constitutively repressesfrdA, and the presence of the linker loop 5-6 reverses this repression [65]. On the other hand, previous DNase I footprint analyses of thenarGpromoter have revealed two areas of protection by NarL-P, with lower affinity centered at the 195 region [15, 66]. It was also shown in a footprint assay that the NarL OD recognizes sites around the narG promoter, and optimal binding activities are induced when the RD (receiver domain) is phosphorylated; therefore, NarL RD phosphorylation acts as an on-off switch for increasing the DNA-binding affinity [65]. Finally, based on an analytical ultracentrifugation study, it was shown that NarL phosphorylation results in RD dimer formation in solution, while the OD is monomeric; both events are completely necessary for stimulating transcription [65]. In unphosphorylated NarL, the DNA-binding site of the output domain is blocked by the receiver domain [61]. Phosphorylation induces the separation of both domains, resulting in the exposure of previously inaccessible DNA-binding residues and the formation of a new dimerization interface in the C-terminal domain [67, 68]. The receiver domain of NarL and NarP could function as a stabilizer of the carboxyl-terminal domain (CTD); it enhances DNA binding and/or enables repression per se in vivo. In addition, the receiver may make protein-protein contacts necessary for transcription activation [69]. NarL (E273 residue) appears to interact directly with the RNA polymerase (α-CTD) to activate the transcription of the ogt promoter [39]. The NarL consensus recognition sequence consists of a 7-bp element, termed the NarL heptamer [11, 28]. A subset of NarL-binding sites, in which two inverted repeats of the heptamer are separated by 2 bp (termed 7-2-7 sites), also represents binding sites for NarP [14, 37]. Therefore, the NarL and NarP regulons overlap. NarL-binding heptamers are found as single copies, inverted repeats, or direct repeats at positions between +20 and -200 relative to the transcriptional start sites [23]. Most NarL-dependent promoters are also codependent on the oxygen sensor FNR, IHF, or Fis, implying that multiprotein complexes are involved in the transcriptional regulation of NarL target promoters [31, 35, 70, 71, 72]. The Nar sites are classified based on the location of the activator binding site (ref). Nar class I promoters have a single Nar-binding site immediately adjacent to the promoter [40, 46, 73], and Nar class II promoters have two Nar-binding sites upstream of the promoter [69] and reference therein. The accumulation of the protein NarL is decreased during iron-limiting conditions [74]. Under ampicillin treatment, the activity of NarL appears to increase [75]. A review of the regulation of molybdenum cofactor (Moco) biosynthesis and molybdoenzyme gene expression via molybdenum and iron by FNR, NarXL/QP, Fur, and ArcA has been published [76]