Antimicrobial Activity of Rhenium Di- and Tricarbonyl Diimine Complexes: Insights on Membrane-Bound S. aureus Protein Binding

Abstract

Antimicrobial resistance is one of the major human health threats, with significant impacts on the global economy. Antibiotics are becoming increasingly ineffective as drug-resistance spreads, imposing an urgent need for new and innovative antimicrobial agents. Metal complexes are an untapped source of antimicrobial potential. Rhenium complexes, amongst others, are particularly attractive due to their low in vivo toxicity and high antimicrobial activity, but little is known about their targets and mechanism of action. In this study, a series of rhenium di- and tricarbonyl diimine complexes were prepared and evaluated for their antimicrobial potential against eight different microorganisms comprising Gram-negative and -positive bacteria. Our data showed that none of the Re dicarbonyl or neutral tricarbonyl species have either bactericidal or bacteriostatic potential. In order to identify possible targets of the molecules, and thus possibly understand the observed differences in the antimicrobial efficacy of the molecules, we computationally evaluated the binding affinity of active and inactive complexes against structurally characterized membrane-bound S. aureus proteins. The computational analysis indicates two possible major targets for this class of compounds, namely lipoteichoic acids flippase (LtaA) and lipoprotein signal peptidase II (LspA). Our results, consistent with the published in vitro studies, will be useful for the future design of rhenium tricarbonyl diimine-based antibiotics.

Keywords: AutoDock; LspA; LtaA; MRSA; S. aureus; antimicrobial; membrane; proteins; rhenium; tricarbonyl.

Scheme 1

Synthetic scheme for the preparation of the complexes investigated in this study. NN = relevant bidentate diamine ligand; L = pyridine (py) or N-methyl imidazole (MeIm). General conditions: i: Et₄NBr, diglyme; ii: NN, ethanol/water, toluene, or CH₂Cl₂; iii: L, methanol or neat L; iv: Br₂, CH₂Cl₂; v: tetrakis(dimethylamino)ethylene, acetonitrile, under N₂; vi: tetrakis(dimethylamino)ethylene, CH₂Cl₂, under N₂. For more details, refer to Section 4.

Figure 1

Structures and codes of the tested Re dicarbonyl (1–5) and tricarbonyl (6–10) complexes.

Figure 2

ORTEP representations of the crystal structures of Re dicarbonyl complexes. Thermal ellipsoids are at 30% probability. Hydrogen atoms are omitted for clarity. Note: the compounds cis-11 and mer-12 co-crystallize in a mixture, where 10 and 11 are also present.

Figure 3

ORTEP representations of the crystal structures of Re tricarbonyl complexes 6–8 and 10. Thermal ellipsoids are at 30% probability. Hydrogen atoms are omitted for clarity.

Figure 4

Structures of previously published active fac-[Re(CO)₃]⁺ complexes. Complex 13 [32]; complex 14 [31]; complexes 15, 16, 18, and 19 [49,50]; complex 17 [48].

Figure 5

(A) Schematic diagram of the lipoteichoic acid synthetic machinery in MRSA with the possible target of the active antimicrobial rhenium complexes. For more details about scheme (A), see [75]. Computer-generated lowest energy pose of the selected complex 19 in the hydrophobic C-terminal pocket of lipoteichoic acid flippase (LtaA): (B) side view; (C) top view; (D) detail of the binding region. In (C,D), the two amino acid residues most likely involved in the H-bonding interactions with 19 are shown in green.

Figure 6

(A) Schematic diagram of the lipoprotein post-translational processing pathway with the possible target of the active antimicrobial rhenium complexes. For more details about scheme (A), see [73]. Computer-generated lowest energy pose of selected complex 19 in lipoprotein signal peptidase II (LspA): (B) side view; (C) top view; (D) detail of the binding region. In (C,D), the amino acid residue most likely involved in the H-bonding interactions with 19 is shown in green.

Figure 7

Gaussian surface representation of the hydrophobicity of (A) 15 and PBP2 (2OLV); (B) 19 and PBP2a (4DKI); (C) 15 and PBP3 (3VSL); and (D) 19 and PBP4 (5TXI). The red-blue color palette changes from hydrophilic blue to hydrophobic red.

Figure 8

Gaussian surface representation of the hydrophobicity of (A) 18 and Lipoteichoic acid synthase (LtaS; 2W5Q); (B) 15 and type-I signal peptidase (SpsB; 4WVJ); (C) 19 and Lipoteichoic acid flippase (LtaA), top view (6S7V); and (D) 19 and Lipoprotein signal peptidase II (LspA; 6RYP). The red-blue color palette changes from hydrophilic blue to hydrophobic red.