Nature-Inspired Compounds Targeting Escherichia coli WrbA as Biofilm-Modulating Agents: Computational Design, Synthesis, and Biological Evaluation

Abstract

Biofilms pose significant challenges in multiple settings due to their resistance to conventional treatments. In this study, we designed and synthesized a novel class of nature-inspired 5,7-dihydroxy-2,2-dimethylchroman-4-one derivatives as binders of WrbA, a potential target for biofilm modulation. Using a structure-based computational approach, a small library of analogs with varied amide moieties was developed and synthesized. The evaluation of their binding affinity to WrbA demonstrated good-to-excellent K_d values, as confirmed by microscale thermophoresis (MST). Antibiofilm assays against Escherichia coli and Staphylococcus aureus revealed different modulating effects on biofilm formation, conceivably linked to ROS production. These findings emphasize the importance of ROS levels in biofilm, as well as the pivotal role of WrbA as a target in its regulation.

Keywords: MM‐GBSA; antibiofilm assay; chroman‐4‐one derivatives; microscale thermophoresis; reactive oxygen species.

Figure 1

Structure of the recurrent 5,7‐dihydroxy‐2,2‐dimethylchroman‐4‐one moiety (in the green square) and three examples of natural compounds selected by the computational protocol described in our previous work [24].

Figure 2

Compound C (light‐blue surface) in complex with WrbA. The flavin mononucleotide (FMN) cofactor is displayed as green sticks. The protein is represented as a molecular surface, color‐mapped based on the electrostatic potential (blue to red for positively to negatively charged areas, respectively).

Figure 3

Predicted binding mode for Compounds 1b–e (panels A–D, respectively) in complex with WrbA, resulting from the MD frames showing the lowest calculated binding free energy values. The secondary structure of WrbA is represented as gray ribbons with some key residues (Trp98, Thr116, and His133) shown as sticks, while FMN is represented as magenta sticks. Hydrogen bonds are represented as yellow dashed lines, π–π stacking as cyan dashed lines, and halogen bonds as purple dashed lines.

Scheme 1

Reagents and conditions: (i) PPA, 1,4‐dioxane, 60°C, 3 h, N₂ atm; (ii) K₂CO₃, CH₃CN, 0°C, 5 h, N₂ atm; (iii) 2.5 M NaOH, EtOH/H₂O, 0°C, 0.5 h; (iv) 1. HATU, DIPEA, DMF, RT, 45 min, N₂ atm; 2. R‐NH₂, RT, overnight, N₂ atm.

Figure 4

Thermal ellipsoid diagram of 4, with the arbitrary atom‐numbering scheme used in the discussion. Ellipsoids represent atomic displacement parameters at the 40% probability level.

Figure 5

(A) Representation of the intermolecular and intramolecular H bonds (green lines). Symmetry‐equivalent interactions are not shown. (B) Representation of the crystal packing, viewed along the a axis. Hydrogen atoms have been omitted for clarity.

Figure 6

(A) HS mapped over d _norm with a fixed color scale from −0.7373 au (red) to 1.1305 au (blue), representing intermolecular contact length relative to the sum of the van der Waals radii (red: shorter; blue: longer; white: equivalent). (B) Two‐dimensional fingerprint plots of the HS, illustrating the frequency of each d _e and d _i combination. Points are color‐coded from blue to green based on their contribution.

Figure 7

MST curves normalized by fraction bound acquired by incubation of recombinant WrbA protein with scaling concentrations of Compounds 1a (red), 1c (orange), 1d (green), and 1e (blue), using a Monolith NT.115 instrument. Two independent experiments were conducted to generate the K _d curve.

Figure 8

Planktonic growth of E. coli (A) and S. aureus (B) in the presence or absence (control) of 500 μM of each compound. Data are expressed as the mean of three independent measurements.