Differential regulation by the homologous response regulators NarL and NarP of Escherichia coli K-12 depends on DNA binding site arrangement

The NarL and NarP proteins are homologous response regulators of Escherichia coli that control the expression of several operons in response to nitrate and nitrite. A consensus heptameric NarL DNA-binding sequence has been identified, and previous observations suggest that the NarP protein has a similar sequence specificity. However, some operons are regulated by NarL alone, whereas others are controlled by both NarL and NarP. In this study, DNase I footprinting experiments with the fdnG , nirB and nrfA control regions revealed that NarP only binds to heptamer sequences organized as an inverted repeat with a 2 bp spacing (7–2–7 sites). The NarL protein also binds to these 7–2–7 sites but, unlike NarP, also recognizes heptamers in other arrangements. These results provide an explanation for the regulation of some operons by NarL alone and for the different effects of NarL and NarP at common target operons, such as fdnG and nrfA . To investigate this differential DNA binding further, derivatives of the nrfA control region were constructed in which the spacing of the 7–2–7 heptamers was increased (7– n –7 constructs). Increasing the spacing to four or more basepairs abolished NarP binding and significantly reduced NarL binding. The NarL protein also had a reduced binding affinity for heptamers adjacent to the 7– n –7 heptamer pair, suggesting a decrease in cooperative interactions. In conclusion, we propose that 7–2–7 sites are preferred by both NarL and NarP. NarL can also recognize other binding site arrangements, an ability that appears to be lacking in NarP.     

Nitrate repression of the Escherichia coli pfl operon is mediated by the dual sensors NarQ and NarX and the dual regulators NarL and NarP.

The pfl operon is expressed at high levels anaerobically. Growth of Escherichia coli in the presence of nitrate or nitrite led to a 45% decrease in expression when cells were cultivated in rich medium. Nitrate repression, however, was significantly enhanced (sevenfold) when the cells were cultured in minimal medium. Regulation of pfl expression by nitrate was dependent on the NarL, NarP, NarQ, and NarX proteins but independent of FNR, ArcA, and integration host factor, which are additional regulators of pfl expression. Strains unable to synthesize any one of the NarL, NarP, NarQ, or NarX proteins, but retaining the capacity to synthesize the remaining three, exhibited essentially normal nitrate regulation. In contrast, narL narP and narX narQ double null mutants were devoid of nitrate regulation when cultured in rich medium but they retained some nitrate repression (1.3-fold) when grown in minimal medium. By using lacZ fusions, it was possible to localize the DNA sequences required to mediate nitrate repression to the pfl promoter-regulatory region. DNase I footprinting studies identified five potential binding sites for the wild-type NarL protein in the pfl promoter-regulatory region. Specific footprints were obtained only when NarL was phosphorylated with acetyl phosphate before the binding reaction was performed. Each of the protected regions contained at least one heptamer sequence which has been deduced from mutagenesis studies to be essential for NarL binding (K. Tyson, A. Bell, J. Cole, and S. Busby, Mol. Microbiol. 7:151-157, 1993).     

Mutants of the lac promoter with large insertions and deletions between the CAP binding site and the -35 region

A set of the lac promoter mutants that have varying lengths of the spacer between the CAP binding site and the -35 region was constructed. The mutants have the spacer length increased by five (I5 mutant), or eleven (I11) residues or decreased by eleven residues (D11). We also present a construction of the hybrid between the gal and lac promoters in which the CAP binding site and the -35 region of the gal promoter are fused to the lac -10 region. The promoter fragments were assembled through ligations of chemically synthesized oligodeoxynucleotides and cloned into a pBR322-derivative vector. The results of the in vivo assays of promoter activity show that the I11 mutation results in an active but weak promoter that can be stimulated by CAP, though to a lesser degree than the wild-type lac promoter. The other mutants exhibit no promoter activity. Since the insertion of 11-bp preserves the location of the CAP binding site on the same side on the DNA helix, the data demonstrate the importance of spatial alignment between the CAP binding site and the promoter. The fact that the gal::lac hybrid is inactive as a promoter indicates also that catabolite activation is a highly complex process in which the -35 and -10 regions cannot be easily exchanged between promoters.