Positive regulation of fur gene expression via direct interaction of fur in a pathogenic bacterium, Vibrio vulnificus

Abstract

In pathogenic bacteria, the ability to acquire iron, which is mainly regulated by the ferric uptake regulator (Fur), is essential to maintain growth as well as its virulence. In Vibrio vulnificus, a human pathogen causing gastroenteritis and septicemia, fur gene expression is positively regulated by Fur when the iron concentration is limited (H.-J. Lee et al., J. Bacteriol. 185:5891-5896, 2003). Footprinting analysis revealed that an upstream region of the fur gene was protected by the Fur protein from DNase I under iron-depleted conditions. The protected region, from -142 to -106 relative to the transcription start site of the fur gene, contains distinct AT-rich repeats. Mutagenesis of this repeated sequence resulted in abolishment of binding by Fur. To confirm the role of this cis-acting element in Fur-mediated control of its own gene in vivo, fur expression was monitored in V. vulnificus strains using a transcriptional fusion containing the mutagenized Fur-binding site (fur(mt)::luxAB). Expression of fur(mt)::luxAB showed that it was not regulated by Fur and was not influenced by iron concentration. Therefore, this study demonstrates that V. vulnificus Fur acts as a positive regulator under iron-limited conditions by direct interaction with the fur upstream region.

FIG. 1.

Effect of iron availability on intracellular levels of Fur protein in wild-type V. vulnificus. V. vulnificus grown in LBS medium at the exponential phase and at the stationary phase were used for estimating cellular contents of Fur protein. Iron in the medium was depleted by adding 0.2 mM 2,2′-dipyridyl to the early exponential cultures (OD₆₀₀ of 0.1). Forty micrograms of each bacterial lysate was fractionated by SDS-PAGE. The blotted membrane was treated with polyclonal antibodies raised against recombinant Fur and then with alkaline phosphatase-conjugated rabbit anti-rat IgG. Upon incubation with the NBT-BCIP system, the Fur protein (17 kDa) appeared as an immunoreactive band, as indicated by the arrow. Lane 1, lysate prepared from exponential phase V. vulnificus in medium without iron chelator; lane 2, lysate prepared from exponential phase V. vulnificus in medium with iron chelator; lane 3, lysate prepared from stationary phase V. vulnificus in medium without iron chelator; and lane 4, lysate prepared from stationary phase V. vulnificus in medium with iron chelator.

FIG. 2.

Effect of the Δhfq mutation on the intracellular levels of RpoS and Fur protein. Lysates of wild type and Δhfq mutant grown at the exponential phase (OD₆₀₀ of 0.5) and at the stationary phase (OD₆₀₀ of 2.0) were used for estimating cellular contents of RpoS (A) and Fur protein (B). Forty micrograms of each bacterial lysate was fractionated by SDS-PAGE. The blotted membrane was treated with polyclonal antibodies raised against recombinant RpoS or Fur and then with alkaline phosphatase-conjugated rabbit anti-rat IgG. Upon incubation with the NBT-BCIP system, the Fur protein or RpoS protein appeared as an immunoreactive band, indicated by an arrow. Lanes 1, protein size marker; lanes 2, wild type at exponential phase; lanes 3, Δhfq mutant at exponential phase; lanes 4, wild type at stationary phase; and lanes 5, Δhfq mutant at stationary phase.

FIG. 3.

Effect of iron availability on fur::luxAB (pHL01) expression in the wild-type and in the Δfur mutant V. vulnificus. Wild type (WT) and Δfur mutant V. vulnificus carrying pHL01 were grown in LBS medium supplemented with 5 μg ml⁻¹ tetracycline. Iron in the medium was depleted by adding 0.2 mM 2,2′-dipyridyl to the early exponential cultures (OD₆₀₀ of 0.1). The fur::luxAB activities were normalized by dividing the number of RLUs by the OD₆₀₀ value. The fur::luxAB activities of four independent cultures at exponential phase (OD₆₀₀ of 0.5) were averaged and are indicated with their standard deviations.

FIG. 4.

Binding of Fur protein to the upstream region of the fur gene. DNase I-footprinting assays were performed to localize the Fur-binding site in the regulatory region of the fur gene in the absence of iron chelator (A) or in the presence of iron chelator (B). The ³²P-labeled 482 bp-DNA fragment of the fur promoter region was incubated with increasing amounts of Fur protein, ranging from 1.5 to 17 μM, and the binding reactions were then treated with DNase I. The reaction mixtures were resolved on a 6% polyacrylamide sequencing gel alongside the sequencing ladder derived from the plasmid pGEMT-fur. The protected region of the fur promoter is shown by a bracket. (A) Lane 1, DNA without Fur; lanes 2 to 4, DNA with recombinant Fur protein at 5.8, 12.0, and 17.0 μM, respectively. (B) Lane 1, DNA without Fur; lanes 2 to 6, DNA with recombinant Fur protein at 1.5, 2.9, 5.8, 12.0, and 17.0 μM, respectively.

FIG. 5.

Upstream region of the fur gene of V. vulnificus MO6-24/O. (A) The putative −10 and −35 sequence of the fur promoter are indicated in bold capitals. The transcriptional initiation site for the fur gene is represented with an arrowhead. Both the ribosomal binding sequence (RBS) and initiation codon (IC) for Fur protein are underlined. The Fur-protected region in the fur promoter is marked in a box and is located from nucleotides −142 to −106 with respect to the transcriptional start site of the fur gene. Mutagenized bases in the mutant fur promoter are indicated above the Fur binding site. (B) Two transcriptional fusions with the wild type (fur::luxAB; pHL01) or mutated Fur binding site (fur_mt::luxAB; pHL03) are represented in a schematic picture. The putative −10 and −35 sequences of the fur promoter are indicated in closed boxes, whereas the transcriptional initiation site for the fur gene is shown as an arrowhead. The Fur-binding region is displayed as an open box, and the altered region in the mutant fur promoter is indicated by gray boxes.

FIG. 6.

Effect of the mutated Fur binding site on Fur-P_fur interaction (A and B) and expression of fur_mt::luxAB (pHL01) and fur_mt::luxAB (pHL03) (C and D). Binding of Fur to the mutated Fur binding site was examined by a DNase I-footprinting assay in the absence of iron chelator in panel A or in the presence of iron chelator in panel B. The ³²P-labeled 482 bp-DNA fragment of the fur promoter region with the mutated Fur binding site was incubated with increasing amounts of Fur protein, ranging from 1.5 to 17 μM, and then treated with DNase I. The reaction mixtures were resolved on a 6% polyacrylamide sequencing gel alongside the sequencing ladder derived from the plasmid pGEMT-fur. (A) Lane 1, DNA without Fur; lanes 2 to 4, DNA with recombinant Fur protein at 5.8, 12.0, and 17.0 μM, respectively. (B) Lane 1, DNA without Fur; lanes 2 to 6, DNA with recombinant Fur protein at 1.5, 2.9, 5.8, 12.0, and 17.0 μM, respectively. Wild-type (C) and Δfur mutant (D) strains carrying pHL01 or pHL03 were grown in LBS medium supplemented with 5 μg ml⁻¹ tetracycline and measured for luciferase activity. Iron in the medium was depleted by adding 0.2 mM 2,2′-dipyridyl, an iron chelator, into the cultures at the time points indicated with arrows. Luciferase activities are expressed as normalized values: number of RLUs divided by the OD₆₀₀ value of each sample. Values are shown for the activity of fur::luxAB (pHL01) in the absence (closed circles) or presence (open circles) of an iron chelator and for fur_mt::luxAB (pHL03) containing the mutated Fur binding site in V. vulnificus strains cultivated without (closed triangles) or with (open triangles) the addition of an iron chelator.