MemProtMD: A resource for membrane proteins

Membrane proteins play key roles in cell biology; as ion channels, drug receptors, and transporters of solute. In addition, ~50% of potential new drug targets are membrane proteins. It has been estimated that a quarter of genes code for membrane proteins. However there is a huge deficit in structural information for these proteins, with membrane proteins currently constituting less than 2% of the known 3D structures. Improvements in the experimental methods for capturing the architecture of membrane proteins has recently lead to a conspicuous rise in resolved structures. Nevertheless these structures are predominantly elucidated either with none or only a few membrane lipids or detergent lipid-mimetics bound. Computer simulations allow us to develop models of the interactions of membrane proteins with lipids or detergents. It is timely therefore, on the 25th anniversary of the determination of the first atomic resolution membrane structure, that a mechanism should be implemented to enable the semi-automation of a high throughput simulation pipeline to provide a clear description of the interactions of a membrane protein structure within its native membrane environment.

For a comprehensive understanding of membrane protein structures knowledge of the protein position within the lipid bilayer is required. As the number of determined structures of membrane proteins increases so does the need for computational methods which predict their position in the lipid bilayer. This proposal outlines a high-throughput pipeline for computer simulations to model the self-assembly of biological membranes around membrane protein structures. An overview of the methods involved is described below: (i) Application of the coarse-grained molecular dynamics (CGMD) simulation approach to lipid bilayer self-assembly around membrane proteins, followed by conversion of the membrane protein-lipid complex to atomistic detail for further molecular simulation. (ii) Development of a high-throughput methodology that enables setup, execution, deposition and analysis. (iii) Creation of lipids and detergent molecule representations for which there are currently no force-field parameters. (iv) Significant improvement and expansion of the current database structure and website front-end. (v) Enhancement of the current analytical tools to incorporate a bioinformatics approach to protein-lipid interactions with a cell membrane.

Who will benefit from this research? 1. The pharmaceutical industry and their stakeholders 2. The bionanotech sector and their stakeholders 3. Regulatory authorities (UK and EU) concerned with environmental exposure to chemicals. How will they benefit from this research? 1. Understanding how membrane proteins interact with their lipid bilayer environment will allow pharmaceutical companies to better design drugs that are transported more readily across such membranes, thereby increasing their bioavailability. It will also enable better targeting of drugs to sites on membrane proteins 'buried' within the bilayer. Both of these aspects should increase the number of compounds that succeed in reaching the market, and thus have both a commercial and a socio-economic impact. 2. The bionanotech sector are interested in membrane proteins in potential biosensors. An example of such a company would be Oxford Nanopore Technologies. Improved understanding of the nature of protein/lipid interactions is likely to facilitate design of more stable membrane proteins for use in nanodevices. 3. An improved knowledge of membrane proteins and their bilayer surroundings will be relevant to e.g. regulatory authorities concerned with environmental exposure to chemicals. Again this will be via improved understanding of how compounds (in this case toxic) may access membrane proteins. Thus, the overall impact will be to advance UK knowledge and technological development and ultimately add to the competitiveness of UK industry.