Characterization of the Zinc Uptake Repressor (Zur) from Acinetobacter baumannii

Abstract

Bacterial cells tightly regulate the intracellular concentrations of essential transition metal ions by deploying a panel of metal-regulated transcriptional repressors and activators that bind to operator-promoter regions upstream of regulated genes. Like other zinc uptake regulator (Zur) proteins, Acinetobacter baumannii Zur represses transcription of its regulon when Zn^II is replete and binds more weakly to DNA when Zn^II is limiting. Previous studies established that Zur proteins are homodimeric and harbor at least two metal sites per protomer or four per dimer. Cd^II X-ray absorption spectroscopy (XAS) of the Cd₂Zn₂ AbZur metalloderivative with Cd^II bound to the allosteric sites reveals a S(N/O)₃ first coordination shell. Site-directed mutagenesis suggests that H89 and C100 from the N-terminal DNA binding domain and H107 and E122 from the C-terminal dimerization domain comprise the regulatory metal site. K_Zn for this allosteric site is 6.0 (±2.2) × 10¹² M^-1 with a functional "division of labor" among the four metal ligands. N-terminal domain ligands H89 and C100 contribute far more to K_Zn than H107 and E122, while C100S AbZur uniquely fails to bind to DNA tightly as measured by an in vitro transcription assay. The heterotropic allosteric coupling free energy, ΔG_c, is negative, consistent with a higher K_Zn for the AbZur-DNA complex and defining a bioavailable Zn^II set-point of ≈6 × 10^-14 M. Small-angle X-ray scattering (SAXS) experiments reveal that only the wild-type Zn homodimer undergoes allosteric switching, while the C100S AbZur fails to switch. These data collectively suggest that switching to a high affinity DNA-binding conformation involves a rotation/translation of one protomer relative to the other in a way that is dependent on the integrity of C100. We place these findings in the context of other Zur proteins and Fur family repressors more broadly.

Figure 1.

Modeling of the AbZur homodimer structure. (A) Alphafold2 model of the complex of AbZur homodimer bound to a DNA containing a single zur box sequence, based on the published crystallographic structure of the E. coli Zur DNA-bound complex. The N-terminal DNA binding domain and the C-terminal dimerization domain are shaded differently in each of the two protomers (blue and green). (B) Close-up of the proposed tetrathiolate structural Zn site 1. (C) Close-up view of the regulatory site 2, with H89 and C100 derived from the DBD and H107 and E122 derived from the dimerization domain. This model has been deposited in Model Archive (https://modelarchive.org/doi/10.5452/ma-tihn6, with associated statistics).

Figure 2.

X-ray absorption spectroscopy of Cd-AbZur. (A) X-ray absorption near-edge spectrum (XANES) revealing Cd-specific excitation. (B) Extended X-ray absorption fine structure (EXAFS) of the Cd coordination complex (left) and the associated FT of these data (right). Black, experimental data; green, fitted data which returns the experimental parameters shown in Table 1.

Figure 3.

Electronic spectroscopy of Co^II-substituted wild-type and C100S AbZur. (A) Titration of Co into 200 μM protomer WT AbZur (gray spectra), up to 1.0 mol equiv (black) in 0.1 mol equiv; blue spectra, up to 2.0 mol equiv (dark blue) in 0.2 mol equiv. Inset, d–d transition region of WT AbZur before and after bleaching by 1 or 2 mol equiv of Zn as indicated. +1 Zn spectra (purple) and +2 Zn spectra (red) are overlaid. (B) Titration of Co into 200 μM C100S AbZur, same titration points. The addition of Zn resulted in the precipitation of the protein. Conditions: 25 mM HEPES, 200 mM NaCl, pH 8.0, and ambient temperature.

Figure 4.

Zn binding to wild-type AbZur. (A) Representative titration of Zn into a solution of quin-2 (3 μM) and apo-AbZur (1 μM). The dashed lines illustrate simulated fits corresponding to 10-fold higher and lower affinities at the binding site. (B) Titration of Zn into apo-AbZur (35 μM dimer) in the presence of 3.0 mM nitriloacetic acid at 15, 25, and 35 °C. Conditions: 25 mM HEPES, 150 mM NaCl, 2 mM TCEP, 3.0 mM NTA, pH 7.4. (C) Representative titration of Zn into a solution of quin-2 (2 μM) and apo-C100S AbZur (2 μM). (D) Representative titration of Zn into a solution of quin-2 (2 μM) and apo-H89A AbZur (2 μM). K_Zn values are compiled in Table 2. The continuous line through the data in panel (B) is a fit to nonequivalent sites binding model with the parameters compiled in Table 3.

Figure 5.

Zn is an allosteric positive regulator of DNA binding by AbZur and is tuned to ≈0.1 pM free Zn. (A) Representative Zur box DNA binding assays carried out with wild-type Zn-Zur (filled circles) and apo-Zur (open circles). The continuous line through the data represents a fit to 1:1 nondissociable homodimer:DNA binding model using DynaFit, with K_DNA,apo of 1.2 × 10⁷ M⁻¹, and K_DNA,Zn of 1.5 × 10⁸ M⁻¹ from three independent experiments. Conditions: 50 mM Tris, 200 mM NaCl, 10 μM NTA, pH 7.4, and 10 nM dsDNA. The vertical arrow represents that change in anisotropy that is reflected in panel (B). (B) Titration of Zn stock solution into 0.02 μM AbZur homodimer, 50 mM Tris, 200 mM NaCl, 2 mM TCEP, 100 μM EDTA and 10 nM DNA with free Zn calculated using the known Zn-EDTA stability constant at pH 7.4 (Table S1).

Figure 6.

Representative in vitro transcription analysis of wild-type and mutant AbZur proteins (0.02 μM dimer) using ROSALIND. (A) Schematic representation of this real-time assay. T7 RNA polymerase (RNAP) binds to its promoter and initiates transcription of an RNA aptamer (3WjdB), which binds the DFHBI-1T ligand and becomes fluorescent. If Zur is capable of binding to its operator (Zur box), then it inhibits transcription of the aptamer, leading to lower fluorescence. (B) Transcription yields obtained for Zn-Zur and apo-Zur relative to a no Zur control. (C) C28S and C100S Zn-Zurs relative to wild-type Zur and no Zur control. (D) Activities of H89A, H107A and E122A Zurs in this assay.

Figure 7.

Small-angle X-ray scattering of the wild-type and C100S AbZur. (A) Normalized scattering curves of the apo- (dark blue) and Zn₂-loaded (light blue) wild-type AbZur and apo- (red) and Zn₂-loaded (pink) C100S AbZur. (B) Pairwise distance distributions for the apo- (dark blue) and Zn₂ loaded (light blue) wild-type AbZur and apo- (red) and Zn₂-loaded (pink) C100S AbZur derived the data shown in panel (A). (C) Dimensionless Kratky plots of apo- (dark blue) and Zn₂- loaded (light blue) wild-type AbZur and apo- (red) and Zn₂-loaded (pink) C100S AbZur, with R_g obtained from a Guinier plot for each species (Figure S4).