A central research theme of the Khalid group research is multiple scale simulation of biological membranes and in particular, bacterial membranes. We are interested in constructing in silico models with in vivo relevance. Traditionally, the chemical and physical complexities of bacterial (and other biological) membranes have been neglected in atomistic and near-atomistic level modelling; We aim to reverse this trend. We have already published one of the most detailed E.coli outer membrane models reported to date and perhaps the most detailed inner membrane model (Piggot et al, 2011, J Phys Chem B). We are now using coarse-grain models to extend the length scales of the systems we can study. To this end, we have developed a protocol to construct spherical vesicles of any given size thus enabling us to simulate systems that are large enough to be studied by experimental methods too, bridging the traditional gap between in silico and in vitro regimes. Concomitant with the bacterial membrane model-building, we have collaborated with Matthieu Chavent (Oxford) to develop novel visualization techniques for large molecular systems. The resulting images have been featured in the EMBO cover contest, the Biomedical Computation review, and have been the recipient of the US Biophysical Society image award in 2013.
Extending beyond the membranes themselves, we have active collaborations various experimental groups in the UK to model the periplasmic layer that separates the two membranes of E.coli. We are now also using our models to study the mechanisms by which antibacterial agents enter the bacteria via interaction with these layers. Some of this work is part of a large, multi-PI, cross institutional grant proposal recently funded by MRC, on which Syma is a co-PI. We are collaborating with experimental and simulation groups from the UK and overseas to study the interaction of the antibacterial and anti-sepsis agent, polymyxin B1 with both bacterial and mammalian membranes (e.g. PLoS Comp.Biol 2015).
Some of our coarse-grain simulations have already pushed the boundaries of the current state-of-the-art in molecular dynamics simulations; with simulations of spherical vesicles composed of a complex mix of lipids giving system sizes of > 1 million particles.
Coarse-Grain simulation of a vesicle that mimics the E.coli outer membrane Khalid et al, Biochemsoc Trans, 2015
Other ongoing work in the area of bacterial membranes includes studying: (i) individual proteins that are known conduits for antibiotics at the atomistic level and quantum mechanical levels; (ii) species-dependent differences in membrane composition and their biomedical implications in the context of antibiotic design and targeting development of resistance to antibiotics; and (iii) developing truly multiscale models that will enable the study of more mesoscopic behaviour such as build-up of biofilms. Towards the latter, we have an active collaboration with Dominic Tildesley (EPFL) through a joint postdoctoral research fellow who is using Dissipative Particle Dynamics (DPD) to simulate large vesicles parameterised from our coarse-grain MD simulations.
Our ultimate aim is to develop a multiscale model that extends from the atomistic regime to the whole cell level.