Rational antibiotic design: in silico structural comparison of the functional cavities of penicillin-binding proteins and ß-lactamases

Abstract

The class of ß-lactam antibiotics has proven highly efficient in targeting bacterial penicillin-binding proteins (PBP) leading to the blocking of the bacterial cell wall synthesis. However, the benefit of these drugs is limited because of bacterial resistance mechanisms; the most widespread resistance involves ß-lactamase enzymes (ßLACT) that inactivate ß-lactam-based molecules. We focused on PBPs and ßLACTs from enterobacteria, and performed a detailed in silico study of PBPs whose inactivation is lethal for the bacteria and of ßLACTs that have a PBP-type catalytic mechanism. The comparison of the sequences and structures of PBPs and ßLACTs shows an almost perfect conservation of the catalytic site, and a high spatial resemblance of the whole functional cavity despite a very low overall sequence identity. Some notable differences in the functional cavity were observed in the vicinity of the catalytic site: four tyrosines are well conserved in the PBPs, whereas the residues occurring at equivalent positions in the ßLACT families present other physicochemical properties. These tyrosines are thus good candidates to be targeted in designing new antibiotic molecules with increased affinity and specificity for PBPs, with the goal of overcoming drug resistance. Our analysis also identified residues that have similar characteristics in most ßLACT families and different properties in PBPs; these are interesting targets for new ligands that specifically inhibit ßLACT proteins. The in silico approach presented here can be extended to other protein systems in view of guiding and improving rational drug design.

Keywords:

• drug design
• ß-Lactam antibiotic resistance
• enterobacteria
• protein sequence comparisons
• protein structure comparisons

Abbreviations:

• PBP: penicillin binding protein
• ßLACT: ß-lactamase
• 3D: 3-dimensional
• PDB: Protein Data Bank
• RMS: root mean square

Acknowledgments

We thank Jean Marc Kwasigroch for help with the computer programs. MNM has a PhD grant from the Belgian Commission for Cooperation and Development (ARES-CCD). MR is Research Director at the Belgian Fund for Scientific Research (FNRS).