Interaction of Bacillus subtilis Fur (ferric uptake repressor) with the dhb operator in vitro and in vivo

Abstract

Bacillus subtilis contains three metalloregulatory proteins belonging to the ferric uptake repressor (Fur) family: Fur, Zur, and PerR. We have overproduced and purified Fur protein and analyzed its interaction with the operator region controlling the expression of the dihydroxybenzoate siderophore biosynthesis (dhb) operon. The purified protein binds with high affinity and selectivity to the dhb regulatory region. DNA binding does not require added iron, nor is binding reduced by dialysis of Fur against EDTA or treatment with Chelex. Fur selectively inhibits transcription from the dhb promoter by sigmaA RNA polymerase, even if Fur is added after RNA polymerase holoenzyme. Since neither DNA binding nor inhibition of transcription requires the addition of ferrous ion in vitro, the mechanism by which iron regulates Fur function in vivo is not obvious. Mutagenesis of the fur gene reveals that in vivo repression of the dhb operon by iron requires His97, a residue thought to be involved in iron sensing in other Fur homologs. Moreover, we identify His96 as a second likely iron ligand, since a His96Ala mutant mediates repression at 50 microM but not at 5 microM iron. Our data lead us to suggest that Fur is able to bind DNA independently of bound iron and that the in vivo role of iron is to counteract the effect of an inhibitory factor, perhaps another metal ion, that antagonizes this DNA-binding activity.

FIG. 1

Purification of Fur and SDS–12% PAGE analysis of fractions from the purification of Fur. Lanes: M, molecular mass standards; 1 and 2, total proteins from uninduced and IPTG-induced BL21(DE3)/pLysE/pHB6505, respectively; 3, pooled fractions from heparin-Sepharose; 4, pooled fractions from Mono-Q; 5, fraction from Superdex-75. Fur runs just below the 21.5-kDa marker, and the Fur doublet is indicated by A and B on the right.

FIG. 2

dhb promoter-operator region. The −35 and −10 regions of the dhb ς^A-dependent promoter (33) are indicated by underlining. The fur box required for iron-mediated repression of dhb transcription (33) is indicated by double underlining, and the start codon (ATG) for dhbA is shown in uppercase letters. The fragment used in the EMSA experiments extends from ∼90 bp upstream of the −35 element (as indicated) to the right bracket. The AlwNI site used to bisect the dhb fragment for EMSA experiments is indicated.

FIG. 3

Specific binding of Fur to the dhb regulatory region. (A) The dhb promoter fragment (50 pM) was cut with AlwNI, end labeled, and incubated with native or EDTA-treated Fur. Lanes: 1, no Fur; 2, 10 nM Fur; 3, 10 nM EDTA-treated Fur. The positions of the unbound fur box and the shifted complex are indicated by unbound and bound, respectively. (B) The dhb promoter fragment (50 pM) was incubated with Fur for 10 min prior to the addition of cold competitor DNA. Lanes: 1, no Fur; 2, 10 nM Fur; 3, 10 nM Fur and 10 nM dhb competitor DNA; 4, 10 nM Fur and 100 nM dhb competitor DNA; 5, 10 nM Fur and 10 nM nonspecific competitor DNA; 6, 10 nM Fur and 100 nM nonspecific competitor DNA. The shifted complex is displaced only by the addition of the specific dhb competitor DNA.

FIG. 4

Effect of added Fe(II) on the affinity of Fur for the dhb promoter region. A curve for the binding of Fur to the dhb promoter fragment in the presence and absence of Fe(II) is shown. Fur (0.5, 1, 2, 4, 8, 16, and 32 nM) in the presence of Fe(II) (■) and Fur (0.06, 0.125, 0.25, 0.5, 1, 2, 4, 8, 16, 32, 64, and 128 nM) in the absence of Fe(II) (●) were incubated with 50 pM end-labeled template and separated by nondenaturing PAGE, and the percentages of bound and unbound DNA were determined by PhosphorImager analysis with the program ImageQuant (Molecular Dynamics, Inc.).

FIG. 5

Fur specifically represses dhb transcription. (A) Addition of Fur before RNAP. The linearized vector (lanes 1 and 3) or the dhb-containing template DNA (lanes 2, 4, 5, 6, 7, 8, and 9) (4 nM) was incubated with 80 nM RNAP for 10 min at 37°C (lanes 1 and 2) or with Fur for 5 min at RT prior to the addition of 80 nM RNAP and incubation for another 5 min at 37°C (lanes 3 to 9). Lanes: 1 and 2, no Fur; 3 and 4, 80 nM Fur; 5, 64 nM Fur; 6, 32 nM Fur; 7, 16 nM Fur; 8, 8 nM Fur; 9, 4 nM Fur. The positions of the dhb transcript and the vector transcript are indicated on the right. (B) Addition of Fur after RNAP. The linearized vector (lane 1) or the dhb-containing template DNA (lanes 2 to 10) (4 nM) was incubated with 80 nM RNAP for 10 min at 37°C (lanes 1 and 2) or preincubated with 80 nM RNAP for 5 min at 37°C prior to the addition of Fur alone (lanes 3, 5, 7, and 9) or Fur and Fe(II) (lanes 4, 6, 8, and 10) and incubation for another 5 min at RT. Lanes: 1 and 2, no Fur; 3, 4 nM Fur; 4, 4 nM Fur and 10 μM Fe(II); 5, 16 nM Fur; 6, 16 nM Fur and 10 μM Fe(II); 7, 64 nM Fur; 8, 64 nM Fur and 10 μM Fe(II); 9, 256 nM Fur; 10, 256 nM Fur and 10 μM Fe(II). The positions of the dhb transcript and the vector transcript are indicated on the right.

FIG. 6

Complementation of fur::kan by the various histidine-to-alanine and cysteine-to-alanine fur mutations. The mutations were inserted at amyE in HB6637 (fur::kan) by a double recombination event, the strains were grown in MM in the absence (dark gray bar) or presence of 5 μM Fe(III) (hatched bar) or 50 μM Fe(III) (light gray bar), and catechol siderophore yields were expressed as OD₅₁₀/OD₆₀₀ ± standard deviations. WT, wild type.

FIG. 7

Fur is destabilized by all four cysteine mutations and two of the histidine mutations. Pulse-labeled fur mutants were lysed, and Fur was immunoprecipitated from the extracts with polyclonal antibodies against E. coli Fur and protein-G agarose beads. The immunoprecipitated proteins in the supernatant were separated by SDS–12% PAGE, and the gel was analyzed by PhosphorImager analysis. Arrows show full-length Fur (upper band) and a Fur degradation product (lower band). WT, wild type.