(192f) Molecular Models to Assess Barrier Properties of Bacterial Cell Walls

The bacterial cell envelope is a multilayered barrier, comprised of cytoplasmic membrane of phospholipids, periplasm of peptidoglycans and the outer membrane of lipopolysaccharides. With rising bacterial resistance it is imperative to develop a molecular understanding of the interactions of antimicrobial molecules with the bacterial cell envelope. The peptidoglycan is the primary constituent of the bacterial cell wall and an important target for antibiotics and antimicrobials particularly in Gram-positive bacteria. Serving as an exoskeleton, the peptidoglycan resists turgor pressure, dictates cell shape and maintains structural rigidity of cells. It is a hetero-polymer consisting of glycan strands and peptide stems. We develop molecular models for peptidoglycans at atomistic as well as coarse-grained scales. The models incorporate the structural features of the cell walls of Gram-positive Staphylococcus aureus (S. aureus) and Gram-negative Escherichia coli (E. coli) bacteria. The peptide orientation, area per disaccharides, cell wall thickness, cavity size distributions, area stretch modulus and bending modulus are assessed and compared across the Gram-positive and Gram-negative strains. The lateral area per disaccharides for the model E. coli cell wall is ~15% higher than that of the monolayer cell wall structure of S. aureus, indicating relatively higher permeability of the cell wall of E. coli. The charge density and electrostatic potential are tenfold higher in magnitude for the E. coli cell wall compared with S. aureus. The interactions of melittin peptides with the model structures are also investigated. Melittin is found to bind preferentially to the D-ALA residues of the peptidoglycan, and the depth of interaction is found to be higher with the cell wall of E. coli compared to Staphylococcal peptidoglycans. The free energies for the passage of antibacterial thymol molecules through the peptidoglycan structures, computed using umbrella sampling molecular dynamics simulations, do not show any appreciable barrier for insertion and translocation of thymol. Our molecular dynamics simulations and the model force fields for peptidoglycans would open up in silico models for the screening and development of novel therapeutics against virulent bacterial infections.