Search

Transcriptional regulatory proteins typically bind specific DNA sequences with approximately 10(3)-10(7)-fold higher affinity than non-specific DNA and this discrimination is essential for their in vivo function. Here we show that the bacterial leucine-responsive regulatory protein (Lrp) does not follow this trend and has a approximately 20-400-fold binding discrimination between specific and non-specific DNA sequences. We suggest that the dual function of Lrp to regulate genes and to organize DNA utilizes this unique property. A approximately 20-fold decrease in binding affinity from specific DNA is dependent upon cryptic binding sites, including the sequence GN(2-3)TTT and A-tracts. Removal of these sites still results in high binding affinity, only approximately 70-fold weaker than that of specific sites. Similar to Lrp's binding of specific sites in the pap and ilvIH promoters, Lrp binds cooperatively to non-specific DNA; thus, protein/protein interactions are important for both specific and non-specific DNA binding. When considering this cooperativity of Lrp binding, the binding selectivity to specific sites may increase to a maximum of approximately 400-fold. Neither leucine nor the pap-specific local regulator PapI alter Lrp's non-specific binding affinity or cooperative binding of non-specific DNA. We hypothesize that Lrp combines low sequence discrimination and relatively high intracellular protein concentrations to ensure its ability to regulate the transcription of specific genes while also functioning as a nucleoid-associated protein. Modeling of Lrp binding data and comparison to other proteins with regulatory and nucleoid-associated properties suggests similar mechanisms.

The structure of Escherichia coli leucine-responsive regulatory protein (Lrp) co-crystallized with a short duplex oligodeoxynucleotide reveals a novel quaternary assembly in which the protein octamer forms an open, linear array of four dimers. In contrast, structures of the Lrp homologs LrpA, LrpC and AsnC crystallized in the absence of DNA show that these proteins instead form highly symmetrical octamers in which the four dimers form a closed ring. Although the DNA is disordered within the Lrp crystal, comparative analyses suggest that the observed differences in quaternary state may arise from DNA interactions during crystallization. Interconversion of these conformations, possibly in response to DNA or leucine binding, provides an underlying mechanism to alter the relative spatial orientation of the DNA-binding domains. Breaking of the closed octamer symmetry may be a common essential step in the formation of active DNA complexes by all members of the Lrp/AsnC family of transcriptional regulatory proteins.

LRP has recently been shown to interact with the regulatory regions of bacterial ribosomal RNA promoters. Here we study details of the LRP-rDNA interaction by gel retardation and high-resolution footprinting techniques. We show that a second regulator for rRNA transcription, H-NS, facilitates the formation of a higher-order LRP-nucleoprotein complex, probably acting transiently as a DNA chaperone. The macromolecular crowding substance ectoine stabilizes the formation of this dynamic complex, while the amino acid leucine, as a metabolic effector, has the opposite effect. DNase I and hydroxyl radical footprint experiments with LRP-DNA complexes reveal a periodic change of the target DNA structure, which implies extensive DNA wrapping reaching into the promoter core region. We show furthermore that LRP binding is able to constrain supercoils, providing a link between DNA topology and regulation. The results support the conclusion that the bacterial DNA-binding protein LRP, assisted by H-NS, forms a repressive nucleoprotein structure involved in regulation of rRNA transcription. The formation of this regulatory structure appears to be directly affected by environmental changes.

The leucine-responsive regulatory protein (Lrp) of Escherichia coli activates expression of a number of operons and represses expression of others. For some members of the Lrp regulon, exogenous leucine mitigates the effect of Lrp, for some it potentiates the effect of Lrp, and for others it has no effect on Lrp action. For the ilvIH operon that we study, Lrp activates expression in vivo and mediates the repression of the operon by exogenous leucine. We studied Lrp-1, a leucine-insensitive variant, to investigate mechanisms by which leucine alters Lrp action as an activator of ilvIH expression. The Asp114Glu change did not have much effect on the amount of total Lrp-1 in cells but decreased the amount of free Lrp-1 two- to threefold. Lrp monomers associate to form octamers and hexadecamers (hexadecamer form predominates at micromolar concentrations; Kd=5.27x10(-8) M), and leucine promotes the dissociation of Lrp hexadecamer to a leucine-bound octamer. By contrast, Lrp-1 exists primarily as an octamer in solution (equilibrium dissociation constant 6.5x10(-5) M) and leucine had little effect on the equilibrium. Thus, the hexadecameric form that Lrp assumes in the absence of DNA is not required for activation of the ilvIH operon. Both leucine and the lrp-1 mutation reduced the apparent affinity of Lrp binding to ilvIH DNA (contains two groups of binding sites separated by 136 bp) but they have different effects on intrinsic binding affinity and binding cooperativity. Whereas leucine reduced intrinsic binding affinities and interactions of Lrps bound at upstream and downstream regions of ilvIH DNA, it increased cooperative dimer-dimer interactions of Lrps bound to two adjacent sites. By contrast, the lrp-1 mutation did not have much effect on intrinsic binding affinities but it decreased cooperative adjacent dimer-dimer interactions and enhanced interactions of Lrps bound at upstream and downstream regions of ilvIH DNA. Our analysis is consistent with the idea that leucine enhances dimer-dimer interactions that contribute to octamer formation, concomitantly reducing dimer-dimer interactions that contribute to the longer range interactions of Lrps that are required for activation of the ilvIH promoter.

Lrp is a global regulatory protein in Escherichia coli that activates expression of more than a dozen operons and represses expression of another dozen. For some operons, exogenous leucine reduces the extent of Lrp action, for others it potentiates the effect of Lrp, and for yet other operons it has no effect. In an effort to understand how leucine affects Lrp-mediated expression, we examined Lrp self-association and the effect of leucine on self-association using light scattering, chemical cross-linking, and analytical ultracentrifugation. The following results were obtained. (i) Lrp self-associates to a hexadecamer and octamer with the predominant species being hexadecamer at microM concentrations. (ii) Lrp undergoes a leucine-induced dissociation of hexadecamer to octamer. (iii) A mutant Lrp lacking 11 amino acid residues at the C terminus does not form higher-order oligomers, suggesting that the C terminus is involved in subunit association. (iv) At nM concentrations, Lrp dissociates to a dimer. It is proposed that leucine regulates the equilibrium between Lrp oligomers and thus Lrp occupancy of sites within different operons, leading to diverse regulatory patterns.

In Escherichia coli, one of the two genes encoding lysyl-tRNA synthetase, lysU, belongs to the regulon controlled by the leucine-responsive regulatory protein (Lrp). To map the site of Lrp action, mutants escaping regulation in rich medium were generated through random mutagenesis of the lysU promoter region. The mutations showed parallel effects on the strength of Lrp-DNA association, as measured in vitro by gel retardation experiments, and on the degree of repression of lysU expression by Lrp in vivo. In addition, DNase I and hydroxyl radical footprinting experiments indicated that several Lrp molecules bind to a DNA region of over 110 bp in a highly cooperative manner. This region, which encompasses the -35 box of the lysU promoter, was the target of all the mutations affecting the strength of the Lrp-DNA association. These mutations are frequently located in short A + T-rich runs distributed along the Lrp binding region with a periodicity of one helix turn. Because we could find such a regular alternance of A + T runs upstream of several other Lrp-regulated genes, we suggest that this pattern is one feature indicative of the binding of Lrp.

The uropathogenic Escherichia coli colonize the host body by attaching themselves to the epithelial cells through the pyelonephritis-associated pili (pap). The expression of the papBA operon is regulated under a reversible phase-variation mechanism, which partitions the population of cells into those that express the pap and others that do not. The two phases of pap expression are the direct consequences of the two distinct DNA-binding modes exhibited by leucine-responsive regulatory protein (Lrp) in the pap promoter region. In the phase-OFF cells, Lrp occupies the binding sites proximal to the transcription start, blocking transcription initiation. In the phase-ON cells, Lrp occupies the binding sites distal to the transcription start and is thought to promote the CAP (catabolite gene activation protein)-directed transcription initiation. Lrp binds to the proximal binding sites more tightly than to the distal sites, and the switching from phase-OFF to phase-ON requires a local co-regulator, PapI. Here, we used PapI and an isolated DNA-binding domain construct of Lrp to show that there is a DNA co-recognition mechanism by which both proteins acquire enhanced affinity to the distal pap site DNA, to which neither of them binds to an appreciable extent without the other. Also, examination of the binding properties of the Lrp DNA-binding domain presented here led us to propose a new sequence alignment of the six pap Lrp-binding sites. New insights into the design of sequences regulating the pap phase variation as revealed by the pap Lrp-binding site sequences are thus defined and discussed.

Escherichia coli DNA adenine methyltransferase (Dam) and Leucine-responsive regulatory protein (Lrp) are key regulators of the pap operon, which codes for the pilus proteins necessary for uropathogenic E. coli cellular adhesion. The pap operon is regulated by a phase variation mechanism in which the methylation states of two GATC sites in the pap regulatory region and the binding position of Lrp determine whether the pilus genes are expressed. The post-replicative reassembly of Dam, Lrp, and the local regulator PapI onto a hemimethylated pap intermediate is a critical step of the phase variation switching mechanism and is not well understood. We show that Lrp, in the presence and in the absence of PapI and nonspecific DNA, specifically protects pap regulatory GATC sites from Dam methylation when allowed to compete with Dam for assembly on unmethylated and hemimethylated pap DNA. The methylation protection is dependent upon the concentration of Lrp and does not occur with non-regulatory GATC sites. Our data suggest that only at low Lrp concentrations will Dam compete effectively for binding and methylation of the proximal GATC site, leading to a phase switch resulting in the expression of pili.

Lrp is a global regulator of metabolism in Escherichia coli that helps cells respond to changes in environmental conditions. The action of Lrp as a transcriptional activator or repressor is sometimes affected when the medium contains exogenous leucine. In this study, we examined the thermodynamics of leucine binding to Lrp and to a leucine response mutant, Lrp-1, and leucine-induced dissociation of Lrp hexadecamer to leucine-bound octamer. The results of dynamic light-scattering and fluorescence measurements suggest that Lrp has two leucine-binding sites, one a high-affinity site and the other a low-affinity site that is coupled to the dissociation reaction. The Gibbs free energy change for leucine binding to the high-affinity site is about −7.0 kcal/mol. Binding of two leucine molecules to low-affinity sites on the hexadecamer or one leucine molecule to one octamer induces the dissociation of hexadecamer to leucine-bound octamer. The Gibbs free energy change for leucine binding to the low-affinity site was estimated to be in the range −4.66 to −5.03 kcal/mol for leucine binding to an octamer or −6.01 to −6.75 kcal/mol for leucine binding to a hexadecamer. The thermodynamic parameters derived from this study were used together with other data to estimate the distribution of free Lrp hexadecamer, octamer, leucine-bound hexadecamer, and leucine-bound octamer in cells. Mathematical modeling, employed to simulate modulation of Lrp action in response to growth conditions, gave results that are consistent with known patterns of Lrp action on different operons.

Lrp is a global regulator of metabolism in Escherichia coli that helps cells respond to changes in environmental conditions. The action of Lrp as a transcriptional activator or repressor is sometimes affected by whether the medium contains exogenous leucine. The abundance of Lrp in cells is relatively high (about 15 microM in monomer), and given the relatively high Lrp binding affinity in vitro for specific binding sites (nanomolar apparent dissociation constants), the expectation is that all binding sites will be saturated with Lrp in vivo. Here we consider the fraction of the total Lrp in cells that is free and the fraction that is bound to DNA. Using minicell-producing strains, we measured the distribution of Lrp between cytoplasm and nucleoid in cells grown under different nutritional conditions and in cells in different phases of growth. In E. coli cells grown in minimal medium to mid-log phase, the ratio of free to DNA-bound Lrp was about 0.67. This ratio decreased about threefold when the cells were grown in minimal medium supplemented with leucine. Our results also confirmed the previous finding that growth rate regulates lrp expression by as much as three to fourfold. Growth rate-regulated lrp expression, along with changes in the extent of non-specific binding, influences the level of free Lrp in vivo over a 16-fold range. We propose that the net effect of these processes is to regulate the relative concentrations of free Lrp hexadecamer and leucine-bound octamer, leading to promoter selection in response to environmental conditions.

Lrp is a global regulatory protein in Escherichia coli that activates expression of more than a dozen operons and represses expression of another dozen. For some operons, exogenous leucine reduces the extent of Lrp action, for others it potentiates the effect of Lrp, and for yet other operons it has no effect. In an effort to understand how leucine affects Lrp-mediated expression, we examined Lrp self-association and the effect of leucine on self-association using light scattering, chemical cross-linking, and analytical ultracentrifugation. The following results were obtained. (i) Lrp self-associates to a hexadecamer and octamer with the predominant species being hexadecamer at microM concentrations. (ii) Lrp undergoes a leucine-induced dissociation of hexadecamer to octamer. (iii) A mutant Lrp lacking 11 amino acid residues at the C terminus does not form higher-order oligomers, suggesting that the C terminus is involved in subunit association. (iv) At nM concentrations, Lrp dissociates to a dimer. It is proposed that leucine regulates the equilibrium between Lrp oligomers and thus Lrp occupancy of sites within different operons, leading to diverse regulatory patterns.

The general stress-induced sigma subunit sigma s of Escherichia coli RNA polymerase is closely related to the vegetative sigma factor sigma 70. In view of their very similar promoter specificity in vitro, it is unclear how sigma factor selectivity in the expression of sigma s-dependent genes is generated in vivo. The csiD gene is such a strongly sigma s-dependent gene. In contrast to sigma s, which is induced in response to many different stresses, csiD, whose expression is driven from a single promoter, is induced by carbon starvation only. To our knowledge, the csiD promoter is the first characterized promoter which is not only exclusively dependent on sigma s-containing RNA polymerase (E sigma s), but also requires an activator, cAMP-CRP. In addition, leucine-responsive regulatory protein (Lrp) acts as a positive modulator of csiD expression. Also in vitro, E sigma s is more efficient than E sigma 70 in csiD promoter binding, open complex formation and run-off transcription, which might be due to the poor match of the csiD -35 region to the sigma 70 consensus and to transcription by E sigma s being less dependent on contacts in this region. By DNase I protection experiments, a cAMP-CRP binding site centered at -68.5 nucleotides upstream of the csiD transcriptional start site was identified. While cAMP-CRP stimulates E sigma 70 binding, it does not promote open complex formation by E sigma 70, but does so in conjunction with E sigma s. With linear templates, cAMP-CRP significantly stimulates E sigma s-mediated in vitro transcription, whereas transcription by E sigma 70 is negligible and hardly stimulated by cAMP-CRP. These findings may reflect different or less stringent positional requirements for an activator site for E sigma s than for E sigma 70, and indicate that cAMP-CRP contributes to sigma factor selectivity at the csiD promoter. In vitro transcription experiments with super-coiled templates, however, revealed significant cAMP-CRP-stimulated transcription also by E sigma 70. Yet, under these conditions, H-NS was found to restore E sigma s specificity by strongly interfering with cAMP-CRP/E sigma 70-dependent transcription. Lrp strongly and cooperatively binds to multiple sites located between positions -14 and -102 (in a way that suggests DNA wrapping around multiple Lrp molecules) and moderately stimulates in vitro transcription, especially with E sigma s. In summary, we conclude that the csiD promoter has an intrinsic preference for E sigma s, but that also protein factors such as cAMP-CRP, Lrp and probably H-NS as well as DNA conformation contribute to its strong E sigma s selectivity. Furthermore, this strong E sigma s preference in combination with a requirement for high concentrations of the essential activator cAMP-CRP ensures csiD expression under conditions of carbon starvation, but not other stress conditions.

The global regulator Lrp (leucine-responsive regulatory protein), in some cases modulated by its co-regulator leucine, has been shown to regulate more than 40 genes and operons in Escherichia coli. Leucine modulates Lrp regulation of leucine-responsive operons. The level of sensitivity of these operons to leucine varies greatly, but the basis for this variation is only partially understood. One operon controlled by Lrp that is relatively insensitive to leucine is gltBDF, which includes genes specifying the large (GltB) and small (GltD) subunits of glutamate synthase. Earlier gel mobility shift assays have demonstrated that Lrp binds to a fragment of DNA containing the gltBDF promoter region. To further define the nature of this Lrp-gltBDF interaction, DNase I footprinting experiments were performed. The results indicate that Lrp binds cooperatively to three sites quite far upstream, spanning the region from -140 to -260 base-pairs relative to the start of transcription. Phased hypersensitivity is observed throughout the entire binding region, suggesting that Lrp bends the DNA. To determine the relative importance of these three sites for the transcriptional activation of gltBDF, a series of site-directed mutations was generated. The effects of these mutations on Lrp binding were determined both by DNase I footprinting and by quantitative mobility shift assays, while their effects on transcription in vivo were examined by measuring beta-galactosidase activity levels of chromosomal gltB::lacZ operon fusions. Our results indicate that all three sites are required for maximal gene expression, as is the proper phasing of the sites with one another and with the start of transcription. Our results suggest that Lrp binds a central palindromic site, interacting predominantly with the major groove of its DNA target, and that additional dimers bind to flanking sites to form a nucleoprotein activation complex.

Lrp (Leucine-responsive regulatory protein) has recently been recognized as a major regulatory protein that controls the expression of many operons in Escherichia coli. Footprinting and gel retardation experiments with DNA from ilvIH, one of the operons controlled positively by Lrp, indicate that Lrp binds to six sites over a 200 base-pair region upstream from the promoter. Binding of Lrp to some of these sites is highly co-operative. We suggest a consensus sequence for Lrp binding based upon a comparison of six binding sites. An analysis of mutants indicates that five out of six binding sites are important for transcription activation and that two or three adjacent Lrp binding sites act synergistically in vivo. The observed synergistic effects in vivo may result from co-operative binding of Lrp to adjacent sites. We propose a model in which multiple binding sites contribute to the formation of a nucleoprotein complex, but only a particular proximal site positions Lrp properly so that it interacts with RNA polymerase.