Kinetic and inhibition studies of dihydroxybenzoate-AMP ligase from Escherichia coli

Abstract

Inhibition of siderophore biosynthetic pathways in pathogenic bacteria represents a promising strategy for antibacterial drug development. Escherichia coli synthesize and secrete the small molecule iron chelator siderophore, enterobactin, in response to intracellular iron depletion. Here we describe a detailed kinetic analysis of EntE, one of six enzymes in the enterobactin synthetase gene cluster. EntE catalyzes the ATP-dependent condensation of 2,3-dihydroxybenzoic acid (DHB) and phosphopantetheinylated EntB (holo-EntB) to form covalently arylated EntB, a product that is vital for the final assembly of enterobactin. Initial velocity studies show that EntE proceeds via a bi-uni-uni-bi ping-pong kinetic mechanism with a k(cat) equal to 2.8 s(-1) and K(m) values of 2.5, 430, and 2.9 microM for DHB, ATP, and holo-EntB-ArCP, respectively. Inhibition and direct binding experiments suggest that, during the first half-reaction (adenylation), DHB binds first to the free enzyme, followed by ATP and the release of pyrophosphate to form the adenylate intermediate. During the second half-reaction (ligation), phosphopantetheinylated EntB binds to the enzyme followed by the release of products, AMP and arylated EntB. Two hydrolytically stable adenylate analogues, 5'-O-[N-(salicyl)sulfamoyl]adenosine (Sal-AMS) and 5'-O-[N-(2,3-dihydroxybenzoyl)sulfamoyl]adenosine (DHB-AMS), are shown to act as slow-onset tight-binding inhibitors of the enzyme with (app)K(i) values of 0.9 and 3.8 nM, respectively. Direct binding experiments, via isothermal titration calorimetry, reveal low picomolar dissociation constants for both analogues with respect to EntE. The tight binding of Sal-AMS and DHB-AMS to EntE suggests that these compounds may be developed further as effective antibiotics targeted to this enzyme.

Figure 1

Kinetic mechanism of EntE. (A) Assays were performed at varying concentrations of DHB (1–10 μM), saturating holo-EntB-ArCP (20 μM), and fixed concentrations of ATP: 150 (●), 200 (○), 350 (▼), 500 (△), and 1000 μM (■). (B) Assays were completed at varying concentrations of DHB (2–60 μM), saturating ATP (5 mM), and varying concentrations of holo-EntB-ArCP: 3 (●), 5 (○), 10 (▼), 20 (△), and 50 μM (■). (C) Assays were performed at varying concentrations of ATP (200–1000 μM), saturating DHB (80 μM), and varying concentrations of holo-EntB-ArCP: 3 (●), 5 (○), 7.5 (▼), 10 (△), and 20 μM (■). Initial velocities were measured spectrophotometrically in 100 mM HEPES pH 7.8 and 10 mM MgCl₂ using the coupled assay as detailed in Materials and Methods. Points are experimental and the lines are global fits of the data to eq 3 for A, and eq 2 for B and C.

Figure 2

Pyrophosphate as a probe of the ping-pong nature of the reaction. (A) Assays were performed at varying concentrations of DHB (2–60 μM), saturating ATP (5 mM), and varying concentrations of holo-EntB-ArCP: 3 (●), 5 (○), 10 (▼), 20 (△), and 50 μM (■). (B) Assays were performed at varying concentrations of DHB (2–60 μM), saturating ATP (5 mM), and varying concentrations of holo-EntB-ArCP: 3 (●), 5 (○), 10 (▼), 20 μM (△) in the presence of 100 μM PP_i. Initial velocities were measured spectrophotometrically in 100 mM HEPES pH 7.8 and 10 mM MgCl₂ using the coupled assay as described in the text. Points are experimental and lines are global fits of the data to eqs 2 and 3 for A and B, respectively.

Figure 3

DHB binds EntE before ATP. (A) Assays were performed at varying concentrations of ATP (0.3 – 1.5 mM), saturating DHB (80 μM), fixed holo-EntB-ArCP (4 μM), and varying concentrations of AMPCPP: 0 (●), 150 (○), 300 (▼), and 500 μM (△). (B) Assays were performed at varying concentrations of DHB (2 – 80 μM), saturating ATP (5 mM), fixed holo-EntB-ArCP (4 μM), and varying concentrations of AMPCPP: 0 (●), 30 (○), and 50 μM (▼). Initial velocities were measured spectrophotometrically in 100 mM HEPES pH 7.8 and 10 mM MgCl₂ using the coupled assay as described in the Materials and Methods. Points are experimental and lines are global fits of the data to eqs 4 and 5 for A and B, respectively.

Figure 4

Isothermal calorimetry profile of EntE (50 μM) with DHB (1 mM). Experiments were performed as described in Materials and Methods. Top: data obtained from automatic injections of 5 μl of DHB. Bottom: the integrated curve showing experimental points (■) and the best fit (−). A fit of the data to a one-set of sites model produced the following values for the binding of DHB to EntE: n = 0.824 ± 0.00512, ΔH = −16500 ± 140 cal/mol, ΔS = −32.2 cal mol⁻¹ K ⁻¹, and K_A = 115000 ± 2800 M⁻¹.

Figure 5

Bisubstrate analogues act as slow-onset inhibitors of EntE. (A) Assays were performed at saturating concentrations of all three substrates: DHB (80 μM), holo-EntB-ArCP (20 μM), and ATP (5 mM), varying, fixed concentrations of Sal-AMS (as labeled in figure, in nM), and 8 nM EntE. (B) Assays were performed as in A, but with varying, fixed concentrations of DHB-AMS (as labeled in figure, in nM). Initial velocities were measured spectrophotometrically in 100 mM HEPES pH 7.8 and 10 mM MgCl₂ using the coupled assay as described in Materials and Methods. Dashed lines represent an average of five traces at a given inhibitor concentration and fits of the data to eq 6 are shown as solid lines. Inset: Plot of k_obs values obtained from eq 6 versus concentration of Sal-AMS (A) or DHB-AMS (B).

Figure 6

Linear dependence of appK_i values at various concentrations of DHB. Symbols are experimentally determined K_i values determined from assays performed at various concentrations of ATP (0.3 – 1.5 mM), different fixed concentrations of DHB (10, 50, and 80 μM), and saturating holo-EntB-ArCP (20 μM) while in the presence of different fixed concentrations of Sal-AMS (0 – 10 nM) and 14 nM EntE. Eq 4 was used to calculate the appK_i value at each concentration of DHB assayed and the line is a fit of the data to eq 9, yielding an intrinsic K_i of 0.47 nM for Sal-AMS.

Figure 7

Representative ITC profiles of (A) EntE (10 μM) with Sal-AMS (150 μM) and salicylic acid (200 mM) and (B) EntE (5 μM) with DHB-AMS (70 μM) and salicylic acid (200 mM). Top: data obtained for automatic injections of 10 μL of Sal-AMS (A) and DHB-AMS (B). Bottom: the integrated curve showing experimental points (■) and the best fit (−).

Scheme 1

E. coli enterobactin biosynthesis occurs via a non-ribosomal peptide synthetase composed of six genes (entA-F).

Scheme 2

Adenylation/Ligation reaction catalyzed by EntE.

Scheme 3

Structures of bisubstrate analogues. (A) 5′-O-[N-(2,3-dihydroxybenzoyl) sulfamoyl]adenosine (DHB-AMS) and (B) 5′-O-[N-(salicyl) sulfamoyl]adenosine (Sal-AMS).

Scheme 4

Two possible kinetic mechanisms can explain non-linear kinetics: one-step slow-association (A) and two-step isomerization (B).

Scheme 5

Proposed Bi Uni Uni Bi Ping-Pong Kinetic Mechanism of EntE.