Multiscale modelling of biological membranes

Lead Research Organisation: University of Southampton
Department Name: School of Chemistry

Abstract

Biological membranes are fundamental structures employed by nature to encapsulate cells. They are dynamic sytems in which proteins float in a sea of lipids. Biomembranes and associated proteins are fundamental components of many phenomena indispensable for life, such as growth, energy storage, and in general information transduction via neural activity. Lipid bilayers, which constitute the backbone structure of any biological membrane, also play increasingly crucial roles in biomedical technology, particularly in drug research and development. A general problem in this field is that experimental evidence is difficult to obtain and be rationalised, due to the small scale and complexity of the systems studied. Computer simulation is an established methodology which can be used to help in the interpretation of experimental data, guide new lines of investigation and ultimately understand biomembrane structures and phenomena with molecular-level resolution. However, standard simulation techniques are computationally expensive, often prohibitively so. To overcome this problem, in recent years much research has been focused on simplified, so-called coarse-grain models; coarse-grain simulations are faster but less accurate than standard simulations. In this proposal I wish to develop further a multiscale methodology, where efficiency is optimised by modelling the bilayer environment at the coarse-grain level, while accuracy is maintained by retaining a standard atomic-level description of membrane-associated proteins and drugs. Preliminary results obtained with this method are very promising, but substantial effort is required to extend it towards applications to the most fundamental yet complex problems of membrane biophysics. The technique would be fast and accurate enough to be capable of shedding light on the currently much debated roles that the interactions between lipids and proteins play in membrane biology. I also propose to employ the techniques developed to areas of biomedical interest, particularly drug design. All drugs must permeate the cell membrane, and most of them target membrane proteins. The mechanisms of drug permeation and action are diffcult to study by experiments or standard simulations; I will apply the multiscale methodology to better understand such phenomena and hence guide the design of improved or novel drugs. Moreover, I propose to investigate problems relevant in nanotechnology. In particular, this work could provide quantitative insights into the controversial issues regarding the interactions between cell membranes and nanomaterials. Such work would be relevant for the promising biomedical application of nanomaterials, for instance as drug delivery agents. I also propose to use our methodology to assess the important issue of toxicity and biocompatibility of nanomaterials.
 
Description Biological membranes are fundamental structures employed by nature to encapsulate cells. They are dynamic sytems in which proteins float in a sea of lipids. Biomembranes and associated proteins are fundamental components of many phenomena indispensable for life, such as growth, energy storage, and in general information transduction via neural activity. Lipid bilayers, which constitute the backbone structure of any biological membrane, also play increasingly crucial roles in biomedical technology, particularly in drug research and development. A general problem in this field is that experimental evidence is difficult to obtain and be rationalised, due to the small scale and complexity of the systems studied. Computer simulation is an established methodology which can be used to help in the interpretation of experimental data, guide new lines of investigation and ultimately understand biomembrane structures and phenomena with molecular-level resolution. However, standard simulation techniques are computationally expensive, often prohibitively so. To overcome this problem, in recent years much research has been focused on simplified, so-called coarse-grain models; coarse-grain simulations are faster but less accurate than standard simulations. In this project, I developed further a multiscale methodology, where efficiency is optimised by modelling the bilayer environment at the coarse-grain level, while accuracy is maintained by retaining a standard atomic-level description of membrane-associated proteins and drugs. Preliminary results obtained with this method had been very promising, but substantial effort was required to extend it towards applications to the most fundamental yet complex problems of membrane biophysics. Such effort has resulted in the development of a new coarse-grain model. This model has been implemented into a popular parallel simulation engine, making it possible to take advantage of state-of-the-art high-performance computing facilities. The technique is now fast and accurate enough to be capable of addressing fundamental problems in membrane biology, such as the much debated roles played by lipids in controlling protein function. For example, the model has been successfully applied to the study of membranes composed of mixtures of different lipid types. Changes in lipid composition were shown to alter fundamental properties which in turn are postulated to affect protein function. The techniques developed were also applied to areas of biomedical interest, particularly drug design. All drugs must permeate the cell membrane, and most of them target membrane proteins. The mechanisms of drug permeation and action are diffcult to study by experiments or standard simulations; the multiscale methodology developed improved our understanding of such phenomena, and could guide the design of improved or novel drugs. There are several future applications for which the methodology developed is ideally suited; these include, for example, the simulation of membrane proteins, as well as modelling interactions between cell membranes and nanomaterial.
Exploitation Route My findings produced novel simulation methodology and results that help rationalise important biological problems for which an understanding is currently lacking. The findings of this project might be taken forward by the membrane biology, pharmaceutical, bionanotechnology, and biomolecular simulation communities, by using them to develop/optimize drugs and drug delivery systems. A flexible and general simulation framework has been developed and made available to the molecular simulation community, which is using it and taking it forward by adapting it to individual needs.
Sectors Chemicals,Digital/Communication/Information Technologies (including Software),Healthcare,Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology

URL http://www.orsi.sems.qmul.ac.uk/
 
Description My findings have been used to conduct new research work at the University of Southampton, where a PhD student and a postdoc have applied and extended the methodology I have developed (this work is ongoing). Moreover, my findings have been used by researchers at the Turin Polytechnic (Italy), which implemented the methodology I developed on new computing technology (graphic processing units - GPU).
First Year Of Impact 2011
Sector Chemicals,Digital/Communication/Information Technologies (including Software),Healthcare,Pharmaceuticals and Medical Biotechnology
Impact Types Cultural

 
Description Archer time access
Amount £47,000 (GBP)
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Department HECBioSim
Sector Academic/University
Country United Kingdom
Start 10/2014 
End 05/2016
 
Description China Scholarship Council
Amount £100,000 (GBP)
Organisation University of Leeds 
Department China Scholarship Council
Sector Academic/University
Country United Kingdom
Start 10/2013 
End 09/2017
 
Description Collaboration with Polito 
Organisation Polytechnic University of Turin
Country Italy 
Sector Academic/University 
PI Contribution I co-supervised a PhD student to provide guidance on the implementation of my method on a new computer system.
Collaborator Contribution They carried out the implementation of my method on their system.
Impact - GPU implementation of molecular dynamics software: https://code.google.com/p/brahms-md/ - Publication: Shkurti, A., M. Orsi, E. Macii , E. Ficarra, A. Acquaviva. Acceleration of Coarse Grain Molecular Dynamics on GPU Architectures, Journal of Computational Chemistry, 34, 803-818 (2013)
Start Year 2009
 
Title BRAHMS molecular simulation 
Description BRAHMS is a molecular dynamics program for the simulation of biological membranes modeled with the ELBA coarse-grain force field. The source code, written in the C language, comprises ~10000 lines, organized in ~20 modules. 
Type Of Technology Software 
Year Produced 2010 
Open Source License? Yes  
Impact BRAHMS has been used in the Essex research group at the University of Southampton and in the Macii group at the Turin Polytechnic. Results obtained with BRAHMS have appeared in several publications: - Shkurti, A., M. Orsi, E. Macii , E. Ficarra, A. Acquaviva. Acceleration of Coarse Grain Molecular Dynamics on GPU Architectures, Journal of Computational Chemistry, 34, 803-818 (2013). - Orsi, M. and J. W. Essex. The ELBA force field for coarse-grain modeling of lipid membranes, PLoS ONE, 6, e28637 (2011) - Orsi, M., M. G. Noro and J. W. Essex. Dual-resolution molecular dynamics simulation of antimicrobials in biomembranes, Journal of the Royal Society Interface, 8, 826-841 (2011). - Orsi, M. and J. W. Essex. Permeability of drugs and hormones through a lipid bilayer: insights from dual-resolution molecular dynamics, Soft Matter, 6, 3797-3808 (2010). - Orsi, M., J. Michel and J. W. Essex. Coarse-grain modelling of DMPC and DOPC lipid bilayers, Journal of Physics: Condensed Matter, 22, 155106 (2010). - Orsi, M., W. E. Sanderson and J. W. Essex. Permeability of small molecules through a lipid bilayer: a multiscale simulation study, Journal of Physical Chemistry B, 113, 12019-12029 (2009). - Orsi, M., D. Y. Haubertin, W. E. Sanderson and J. W. Essex. A quantitative coarse-grain model for lipid bilayers, Journal of Physical Chemistry B, 112, 802-815 (2008). As at Wed 19 Nov 2014, BRAHMS has been donwladed ~ 300 times (https://code.google.com/p/brahms-md/downloads/list). 
URL https://code.google.com/p/brahms-md/
 
Title ELBA-LAMMPS 
Description The ELBA-LAMMPS toolkit is a software package containing tools and examples to facilitate the simulation of the ELBA force field with the massively-parallel molecular dynamics software LAMMPS. 
Type Of Technology Software 
Year Produced 2011 
Open Source License? Yes  
Impact The ELBA-LAMMPS software is routinely used in the Essex group at the University of Southampton as well as in my group (Orsi) at Queen Mary University of London. Results obtained using the ELBA-LAMMPS software have been reported in several publications: - Ding, W., M. Palaiokostas, M. Orsi. Stress testing the ELBA water model, Molecular Simulation, (2016), 42, 337. - Orsi, M., W. Ding, M. Palaiokostas. Direct Mixing of Atomistic Solutes and Coarse-Grained Water, Journal of Chemical Theory and Computation, (2014), 10, 4684. - Orsi, M. Comparative assessment of the ELBA coarse-grained model for water, Molecular Physics, 112, 1566-1576 (2014) - Orsi, M. and J. W. Essex. Physical properties of mixed bilayers containing lamellar and nonlamellar lipids: insights from coarse-grain molecular dynamics simulations, Faraday Discussions, 161, 249-272 (2013) 
URL https://github.com/orsim/elba-lammps