Discovery of Fur binding site clusters in Escherichia coli by information theory models

Fur is a DNA binding protein that represses bacterial iron uptake systems. Eleven footprinted Escherichia coli Fur binding sites were used to create an initial information theory model of Fur binding, which was then refined by adding 13 experimentally confirmed sites. When the refined model was scanned across all available footprinted sequences, sequence walkers, which are visual depictions of predicted binding sites, frequently appeared in clusters that fit the footprints (~83% coverage). This indicated that the model can accurately predict Fur binding. Within the clusters, individual walkers were separated from their neighbors by exactly 3 or 6 bases, consistent with models in which Fur dimers bind on different faces of the DNA helix. When the E. coli genome was scanned, we found 363 unique clusters, which includes all known Fur-repressed genes that are involved in iron metabolism. In contrast, only a few of the known Fur-activated genes have predicted Fur binding sites at their promoters. These observations suggest that Fur is either a direct repressor or an indirect activator. The Pseudomonas aeruginosa and Bacillus subtilis Fur models are highly similar to the E. coli Fur model, suggesting that the Fur–DNA recognition mechanism may be conserved for even distantly related bacteria.     

Coordinate regulation of siderophore and exotoxin A production: molecular cloning and sequencing of the Pseudomonas aeruginosa fur gene.

A 5.9-kb DNA fragment was cloned from Pseudomonas aeruginosa PA103 by its ability to functionally complement a fur mutation in Escherichia coli. A fur null mutant E. coli strain that contains multiple copies of the 5.9-kb DNA fragment produces a 15-kDa protein which cross-reacts with a polyclonal anti-E. coli Fur serum. Sequencing of a subclone of the 5.9-kb DNA fragment identified an open reading frame predicted to encode a protein 53% identical to E. coli Fur and 49% identical to Vibrio cholerae Fur and Yersinia pestis Fur. While there is extensive homology among these Fur proteins, Fur from P. aeruginosa differs markedly at its carboxy terminus from all of the other Fur proteins. It has been proposed that this region is a metal-binding domain in E. coli Fur. A positive selection procedure involving the isolation of manganese-resistant mutants was used to isolate mutants of strain PA103 that produce altered Fur proteins. These manganese-resistant Fur mutants constitutively produce siderophores and exotoxin A when grown in concentrations of iron that normally repress their production. A multicopy plasmid carrying the P. aeruginosa fur gene restores manganese susceptibility and wild-type regulation of exotoxin A and siderophore production in these Fur mutants.     

The gonococcal fur regulon: identification of additional genes involved in major catabolic,recombination, and secretory pathways

In this study, we have characterized the in vitro binding of Neisseria gonorrhoeae Fur to several well-defined iron transport genes, as well as to additional genes involved in major catabolic, secretory, and recombination pathways of gonococci. The gonococcal Fur protein was recombinantly expressed in Escherichia coli HBMV119. Fur was isolated from inclusion bodies and partially purified by ion-exchange chromatography. Gonococcal Fur was found to bind to the promoter/operator region of a gene encoding the previously identified Fur-regulated periplasmic binding protein (FbpA) in a metal ion-dependent fashion, demonstrating that purified Fur is functional. In silico analysis of the partially completed gonococcal genome (FA1090) identified Fur boxes in the promoters of several genes, including tonB, fur, recN, secY, sodB, hemO, hmbR, fumC, a hypothetical gene (Fe-S homolog), and the opa family of genes. By using purified gonococcal Fur, we demonstrate binding to the operator regions of tonB, fur, recN, secY, sodB, hemO, hmbR, fumC, the Fe-S homolog gene, and the opa gene family as determined by an electrophoretic mobility shift assay. While gonococcal Fur was demonstrated to bind to the promoter regions of all 11 opa genes (opaA through -K), we did not detect binding of purified E. coli Fur with 8 of the 11 opa members, indicating that target DNA sequence specificities between these two closely related proteins exist. Furthermore, we observed differences in the relative strengths of binding of gonococcal Fur for these different genes, which most likely reflect a difference in affinity between gonococcal Fur and its DNA targets. This is the first report that definitively demonstrates the binding of gonococcal Fur to its own promoter/operator region, as well as to the opa family of genes that encode surface proteins. Our results demonstrate that the gonococcal Fur protein binds to the regulatory regions of a broad array of genes and indicates that the gonococcal Fur regulon is larger than originally proposed.     

Ferric uptake regulator protein: binding free energy calculations and per-residue free energy decomposition

Iron homeostasis is, in many bacterial species, mediated by the ferric uptake regulator (Fur). A regulatory site able to bind iron to activate Fur for DNA binding has been described, and a structural zinc site essential for the dimerization has also been proposed. They have been localized and named site 1 and site 2, respectively, from the crystal structure of a zinc-substituted Pseudomonas aeruginosa Fur (PA-Fur). Notwithstanding the studies on Fur proteins from various species, both the precise site of iron binding and the effect on DNA binding affinity are still controversial. These issues were investigated here by molecular dynamics simulations and free energy calculations. Simulations were performed for eight molecular systems represented by the three forms of Fur, that is, apo Fur, metal-substituted Fur, and Fur complexed with DNA. Because of the lack of a Fur-DNA complex crystal structure, the recently published model based on mass spectrometry experiments on Escherichia coli Fur (EC-Fur), and the crystal structure of PA-Fur, was used, after adjustment to adopt a symmetric conformation. The simulation results suggest that the formerly proposed site 2 is, in fact, the regulatory iron-sensing site. The calculations also predict that Fe(2+) at site 2 is hexacoordinated having an octahedral environment with only nitrogen and oxygen atoms, which is in accordance with previous spectroscopic characterizations. Energy decomposition pinpoints H87 as an additional amino acid that defines the regulatory metal site. Finally, free energy decomposition analysis reveals a number of amino acids potentially important in dimerization and in DNA binding.     

Architecture of a protein central to iron homeostasis: crystal structure and spectroscopic analysis of the ferric uptake regulator

Iron is an essential element for almost all organisms, although an overload of this element results in toxicity because of the formation of hydroxyl radicals. Consequently, most living entities have developed sophisticated mechanisms to control their intracellular iron concentration. In many bacteria, including the opportunistic pathogen Pseudomonas aeruginosa, this task is performed by the ferric uptake regulator (Fur). Fur controls a wide variety of basic physiological processes including iron uptake systems and the expression of exotoxin A. Here, we present the first crystal structure of Fur from P. aeruginosa in complex with Zn2+ determined at a resolution of 1.8 A. Furthermore, X-ray absorption spectroscopic measurements and microPIXE analysis were performed in order to characterize the distinct zinc and iron binding sites in solution. The combination of these complementary techniques enables us to present a model for the activation and DNA binding of the Fur protein.     

Structural dynamics and functional domains of the fur protein

Proteolytic enzymes were used to detect metal-induced conformational changes in the ferric uptake regulation (Fur) protein of Escherichia coli K12. Metal binding results in enhanced cleavage of the N-terminal region of Fur by trypsin and chymotrypsin. Activation of both trypsinolysis sensitivity and DNA binding have similar metal ion specificity and concentration dependencies, suggesting that the conformational change detected is required for operator DNA binding. Isolation and characterization of biochemically generated fragments of Fur as well as other data indicate that the N-terminal region is necessary for the interaction of the repressor with DNA and that a C-terminal domain is sufficient for binding to metal ions.     

Recognition of DNA by Fur: a reinterpretation of the Fur box consensus sequence

Ferric uptake repressor (Fur) proteins regulate the expression of iron homeostasis genes in response to intracellular iron levels. In general, Fur proteins bind with high affinity to a 19-bp inverted repeat sequence known as the Fur box. An alignment of 19 operator sites recognized by Bacillus subtilis Fur revealed a different conserved 15-bp (7-1-7) inverted repeat present twice within this 19-bp consensus sequence. We demonstrated using electrophoretic mobility shift assays that this 7-1-7 inverted repeat comprises a minimal recognition site for high-affinity binding by Fur. The resulting revised consensus sequence is remarkably similar to a related 7-1-7 inverted repeat sequence recognized by PerR, a Fur paralog. Our analysis of the affinity and stoichiometry of DNA binding by B. subtilis Fur, together with a reinterpretation of previously described studies of Escherichia coli Fur, supports a model in which the 19-bp Fur box represents overlapping recognition sites for two Fur dimers bound to opposite faces of the DNA helix. The resulting recognition complex is reminiscent of that observed for the functionally related protein DtxR. Like Fur, DtxR contains a helix-turn-helix DNA-binding motif, recognizes a 19-bp inverted repeat sequence, and has a typical DNase I footprint of approximately 30 bp. By envisioning a similar mode of DNA recognition for Fur, we can account for the internal symmetries noted previously within the Fur box, the tendency of Fur to extend into adjacent regions of DNA in a sequence-selective manner, and the observed patterns of DNA protection against enzymatic and chemical probes.     

Characterization of the DNA- and metal-binding properties of Vibrio anguillarum fur reveals conservation of a structural Zn(2+) ion

The ferric uptake regulator, Fur, represses iron uptake and siderophore biosynthetic genes under iron-replete conditions. Here we report in vitro solution studies on Vibrio anguillarum Fur binding to the consensus 19-bp Escherichia coli iron box in the presence of several divalent metals. We found that V. anguillarum Fur binds the iron box in the presence of Mn(2+), Co(2+), Cd(2+), and to a lesser extent Ni(2+) but, unlike E. coli Fur, not in the presence of Zn(2+). We also found that V. anguillarum Fur contains a structural zinc ion that is necessary yet alone is insufficient for DNA binding.     

Characterization of the DNA-binding site in the ferric uptake regulator protein from Escherichia coli by UV crosslinking and mass spectrometry

Ferric uptake regulator protein (Fur) is activated by its cofactor iron to a state that binds to a specific DNA sequence called 'Fur box'. Using mass spectrometry-based methods, we showed that Tyr 55 of Escherichia coli Fur, as well as the two thymines in positions 18 and 19 of the consensus Fur Box, are involved with binding. A conformational model of the Fur-DNA complex is proposed, in which DNA is in contact with each H4 [A52-A64] Fur helix. We propose that this interaction is a common feature for the Fur-like proteins, such as Zur and PerR, and their respective DNA boxes.     

Identification and cloning of a fur regulatory gene in Yersinia pestis.

Yersinia pestis is one of many microorganisms responding to environmental iron concentrations by regulating the synthesis of proteins and an iron transport system(s). In a number of bacteria, expression of iron uptake systems and other virulence determinants is controlled by the Fur regulatory protein. DNA hybridization analysis revealed that both pigmented and nonpigmented cells of Y. pestis possess a DNA locus homologous to the Escherichia coli fur gene. Introduction of a Fur-regulated beta-galactosidase reporter gene into Y. pestis KIM resulted in iron-responsive beta-galactosidase activity, indicating that Y. pestis KIM expresses a functional Fur regulatory protein. A cloned 1.9-kb ClaI fragment of Y. pestis chromosomal DNA hybridized specifically to the fur gene of E. coli. The coding region of the E. coli fur gene hybridized to a 1.1-kb region at one end of the cloned Y. pestis fragment. The failure of this clone to complement an E. coli fur mutant suggests that the 1.9-kb clone does not contain a functional promoter. Subcloning of this fragment into an inducible expression vector restored Fur regulation in an E. coli fur mutant. In addition, a larger 4.8-kb Y. pestis clone containing the putative promoter region complemented the Fur- phenotype. These results suggest that Y. pestis possesses a functional Fur regulatory protein capable of interacting with the E. coli Fur system. In Y. pestis Fur may regulate the expression of iron transport systems and other virulence factors in response to iron limitation in the environment. Possible candidates for Fur regulation in Y. pestis include genes involved in ferric iron transport as well as hemin, heme/hemopexin, heme/albumin, ferritin, hemoglobin, and hemoglobin/haptoglobin utilization.     

Vibrio cholerae fur mutations associated with loss of repressor activity: implications for the structural-functional relationships of fur.

We used the Vibrio cholerae Fur protein as a model of iron-sensitive repressor proteins in gram-negative bacteria. Utilizing manganese mutagenesis, we isolated twelve independent mutations in V. cholerae fur that resulted in partial or complete loss of Fur repressor function. The mutant fur genes were recovered by PCR and sequenced; 11 of the 12 contained point mutations (two of which were identical), and one contained a 7-bp insertion that resulted in premature truncation of Fur. All of the mutants, except that containing the prematurely truncated Fur, produced protein by Western blot (immunoblot) analysis, although several had substantially smaller amounts of Fur and two made an immunoreactive protein that migrated more rapidly on sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Nine of the 11 point mutations altered amino acids that are identical in all of the fur genes sequenced so far, suggesting that these amino acids may play important structural or functional roles in Fur activity. Eight of the point mutations occurred in the amino-terminal half of Fur, which is thought to mediate DNA binding; most of these mutations occurred in conserved amino acids that have been previously suggested to play a role in the interaction between adjacent alpha-helices of the protein. Three of the point mutations occurred in the carboxy-terminal half of Fur, which is thought to bind iron. One mutation at histidine-90 was associated with complete loss of Fur function; this amino acid is within a motif previously suggested as being involved in iron binding by Fur. The fur allele mutant at histidine-90 interfered with iron regulation by wild-type fur in the same cell when the mutant allele was present at higher copy number; wild-type fur was dominant over all other fur mutant alleles studied. These results are analyzed with respect to previous models of the structure and function of Fur as an iron-sensitive repressor.