Phytochemicals increase the antibacterial activity of antibiotics by acting on a drug efflux pump.
Bottom Line: In silico screening was used to predict the bioactivity of plant compounds and to compare that with the known EPI, phe-arg-β-naphthylamide (PAβN).Subsequently, promising products have been tested for their ability to inhibit efflux.We demonstrated the feasibility of in silico screening to identify compounds that potentiate the action of antibiotics against drug-resistant strains and which might be potentially useful lead compounds for an EPI discovery program.
Affiliation: Department of Pharmacology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1PD, United Kingdom.Show MeSH
Related in: MedlinePlus
Mentions: The binding of natural products to AcrB was predicted through docking experiments and compared with the docking of the known inhibitor PAβN and substrate minocycline. The T (tight or binding) monomer of AcrB (PDB code 4DX5) (Eicher et al. 2012) with minocycline bound to the deep binding pocket was used in the docking experiments. PAβN docked in a position closer to the lateral binding site of the T monomer compared to minocycline (Fig. 3A) and is stabilized by interactions with amino acids Phe-628, Phe-178, Tyr-287, and a hydrogen bond to the oxygen of Gln176 (Fig. 3A). The predicted docking site of PAβN correlates well with the binding of PAβN to AcrB as predicted by dynamic simulations where Phe-628, Phe-178, and Gln-176 were also some key residues for interaction with PAβN (Vargiu and Nikaido 2012). In addition, a recent crystal structure of AcrB and MexB revealed that Phe 178 contributes to the tight binding of an inhibitor molecule through π–π interactions (Nakashima et al. 2013). Incidentally, all the natural products which were predicted to be potential EPIs docked in a similar position as PAβN, while substrates such as minocycline, acridine, and erythromycin docked in the deep binding pocket (Fig. 3). Vargiu and Nikaido (Vargiu and Nikaido 2012) utilized molecular dynamics simulations which also predicted that residues from the deep binding pocket were much more involved in interaction with a range of substrates, while less involved in stabilizing the inhibitor PAβN. In our in silico binding predictions, all natural products were stabilized by hydrogen bonding and extensive hydrophobic interactions between phenylalanine side chains and aromatic groups within the compounds. This is not surprising as phenylalanine residues play an important role in substrate recognition and stabilization in RND-type drug transporters (Ohene-Agyei et al. 2012). According to the docking, mangiferin would act more like a substrate than an inhibitor and the predicted binding affinity was also the lowest of the natural products included in the assays. The free energy of interaction (ΔGbinding) determined from the docking of each compound was used to predict the Ki for the interactions (Table 1). Accordingly, the compounds could be arranged into an order of putative inhibition efficiency given as:
Affiliation: Department of Pharmacology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1PD, United Kingdom.