Developing a Coarse-Grained Model for Bacterial Cell Walls: Evaluating Mechanical Properties and Free Energy Barriers

Understanding bacterial cell wall mechanics is essential for drug design, but atomistic simulations are limited by computational cost. Prior research has shown that Gram-negative cell envelopes have multiple layers, including peptidoglycan. However, no coarse-grained model of peptidoglycan exists in the MARTINI framework. This gap motivated the development of a CG model to enable larger-scale simulations. Existing models for LPS and lipids are available, but peptidoglycan remains unmodeled. The lack of a CG peptidoglycan model limits the ability to study its mechanical and permeability properties. Computational methods like MARTINI allow for longer simulations but require accurate parameterization. This paper addresses the absence of a suitable CG representation for peptidoglycan in bacterial membranes.

The aim of this study is to develop a coarse-grained model of the peptidoglycan network in Gram-negative bacteria using the MARTINI framework. The specific problem is the absence of a CG model for peptidoglycan that can be used in molecular dynamics simulations. This limitation hinders the study of cell wall mechanics at relevant time and length scales. The motivation is to enable more efficient simulations of drug interactions with bacterial membranes. The study focuses on structural and mechanical properties of peptidoglycan. The authors propose to use an all-atom model as a reference for parameterization. The goal is to create a model that reproduces key properties like glycan strand distances and bending modulus. The study also evaluates the model's ability to simulate small molecule insertion into the cell wall.

The researchers used an all-atom model of peptidoglycan described by Gumbart et al. to develop a coarse-grained representation within the MARTINI framework. Structural parameters like end-to-end distance of glycan strands were matched between atomistic and CG models. The equilibrium angle between adjacent peptides was calculated to assess structural accuracy. Area per disaccharide and cavity size distributions were compared to validate the model. Mechanical properties such as area compressibility and bending modulus were evaluated. Umbrella sampling was used to calculate the free energy of thymol insertion into the peptidoglycan network. Restraint-free simulations were performed to observe thymol translocation across the model. The CG model was tested for its ability to reproduce atomistic simulation results while maintaining computational efficiency.

The CG model of peptidoglycan successfully reproduced the end-to-end distance of glycan strands from atomistic simulations. The equilibrium angle between adjacent peptides matched closely with atomistic results. Area per disaccharide and cavity size distributions were consistent with all-atom data. Mechanical properties like area compressibility and bending modulus were accurately captured. The free energy of thymol insertion was less than kBT in both CG and atomistic models. Restraint-free simulations showed rapid translocation of thymol across the peptidoglycan layer. The CG model maintained structural and mechanical accuracy while allowing larger-scale simulations. The results suggest that the model can be used to study drug permeation and cell wall mechanics at longer time scales.

The authors propose that the developed CG model of peptidoglycan is suitable for studying bacterial cell wall mechanics at larger length and time scales. The model reproduces key structural and mechanical properties from atomistic simulations. The free energy of thymol insertion was found to be less than kBT in both models. The CG model enables efficient simulations without sacrificing accuracy. The study suggests that the model can be used to assess drug interactions with bacterial membranes. The authors emphasize the importance of accurate CG models for overcoming computational limitations. The results support the use of the model in future studies of cell wall permeability and mechanical behavior. The model is expected to facilitate the development of novel antibiotics targeting Gram-negative bacteria.