News and Blog

EPSRC CDT in Next Generation Computational Modelling

Can simulations can save us all from superbugs?

Presentation by Professor Syma Khalid

Syma Khalid is a professor of computational biophysics at Southampton. She spoke to us on the 29th of May 2018 about the simulations constructed by her group with the aim of understanding bacterial cell walls.

The discovery of antibiotics in the 20th century revolutionized medicine at the time, but we are now reaching a point where strains of bacteria are becoming more and resistant to many types of antibiotics. This resistance poses a serious problem for the future of the human race, as we may one day reach a point where many infections can no longer be treated with antibiotics. It is therefore vital that we seek to understand how to destroy bacteria more effectively, and this starts with a better understanding of their cell walls, a bacterium’s first line of defence. A basic representation of a bacterial cell wall is show below.

A Bacterial Cell Wall.

A Bacterial Cell Wall

This consists of three layers; the outermost layer that is oil based, a central water based layer gives structure to the cell wall and an innermost later that is also oil based. To allow transport through these layers, the cell produces tunnel like protein structures.

Typical antibiotics are capable of penetrating these layers and destroying the cell. Strains of bacteria with an increased resistance to this process are being observed, these are known as superbugs. One of the mechanisms under investigation for this is an increased quantity of Efflux pumps in the cell wall. Efflux pumps are protein tubes as shown in figure [1], and the hypothesis is that they expel antibiotics from the bacteria.

It is incredibly difficult to fully model the complexity of a biological system. The human body alone contains approximately 1027 atoms. Bacteria only possess a single cell, but that alone can still contain as many as 1010 atoms. This is still far too many to fully model, even on the most powerful supercomputers. Thankfully, the element of interest is the cell wall; Syma and her team at Southampton simulate a section of the cell wall with as many as a million particles.

In 2011, Syma’s team successfully modelled the proteins that make up the outer membrane of the infamous E-Coli bacterium. This model is fully atomistic, individually modelling each atom in the protein (Piggot et al. J Phys Chem B, 2011).

This method is not suitable for all situations, however. When larger scale simulations are required, it is often necessary to use a technique called course graining. Course graining approximates groups of atoms as a single particle, allowing for larger scales to be explored. Using this method, the simulation scales can approach the experimental scales. If experiments can be conducted on the same scales as the simulation, then it is possible to verify the simulation. Verification leads to confidence in the simulation.

Using the simulation, the next step is to observe the interaction of drugs with the cell wall. Understanding this mechanism may improve our current picture of the interactions between antibiotics and bacterial cell walls, and it may also lead to new methods for breaching the cell wall and destroying the bacteria. Researchers hope that future methods, if managed well, will not be defeated by new superbugs.