Simulation of Biological Ion Channels
The long term goal of this research is to provide a comprehensive physical description of biological ion channels. Such a theoretical model, once successfully formulated, will be capable of predicting channel conductance from channel structure, and capable of revealing certain aspects of the atomic structure of protein macromolecules from observed conductance behavior. It will link the structure and function of ion channels through the details of the inter-molecular potential operating between ions, water molecules and atoms that form the channel. This study is aimed at elucidating how anions in moving across chloride channels interact with the induced and fixed charges on the protein wall, with water molecules and other ions in the electrolyte solution.
|
Principal Investigator Shin-Ho ChungTheoretical Physics, RSPhysSE Australian National University |
Project d32 |
|
Co-Investigators Harold SchranzSchool of Chemistry ADFA Turgut Bastug Physics/Science University of Sydney Roger Brown Stephen McMahon ANUSF and ITS Staff, DOI Australian National University Matthew Hoyles Chemistry, Faculty of Science Australian National University Jian Yin Taira Vora David Bisset Megan O'Mara Theoretical Physics, RSPhysSE Australian National University Cheng Chen Paul Altin Patrick Scott Rachel Sarah Blakers Alexander Bissember Physics and Theoretical Physics, Faculty of Science Australian National University Ben Corry Biomedical, Biomolecular and Chemical Science University of Western Australia Toby Allen Biochemistry / Physiology and Biophysics Weill Medical College of Cornell University |
RFCD Codes 270104 |
Significant Achievements, Anticipated Outcomes and Future Work
Our research effort in the past 12 months has been centred on two projects. First, we constructed all-atom homology models of two types of anion-selective, ligand-gated ion channels, namely, the glycine and GABAA receptors. The properties of these receptor models were then rigorously tested to ensure that the models replicate in great fidelity the experimental observations. We devised, to further examine the homology models, two novel computational tools, namely, (i) a learning-based, dynamic control algorithm for simulating biological channels, which we call adaptive controlled Brownian dynamics, and (ii) a hybrid of molecular dynamics and stochastic dynamics algorithms that explicitly represent water molecules and allow for the thermal and induced motions of residues lining the ion- conducting pathway. Among the properties we examined using these newly-devised algorithms are: channel conductances, the current-voltage-concentration profiles, and detailed interactions between the permeant ion with water molecules and the dielectric boundary as it traverses the channel conduit. We then made a set of testable predictions. Second, we created an open-state conformation of the Kv1.2 potassium channel, whose crystal structure has recently been partially determined. The all-atom, open-state structure of the voltage-gated channel so constructed was initially tested, using Brownian dynamics, to ensure that it can accurately replicate the macroscopic observables. We are now planning to elucidate the precise conformational changes the protein undergoes as the channel makes a transition from the open-state to the closed-state, and vice versa. Because such conformational changes take place in the time-scale of 100s of microseconds, it will not be possible to simulate the structure for such a long period using conventional molecular dynamics calculations. To circumvent this difficulty and to make the computation tractable, we will design a special molecular dynamics algorithm that can increase the computational speed by at least 2 orders of magnitude.
Data Sources, Curation Techniques, Data Access Policy and Method
The data we store in the MDSS are mainly the trajectory files created by molecuar dynamics calculations and Brownian dynamics simulations. In general, the positions of atoms, ions and water molecules comprising the simulation assembly are recorded at each discrete time-step for the entire simulation periods, and then the files are compressed before we store in the MDSS.
Our stored data are made available to researchers in the field for further detailed analysis. Upon request, we usually copy the relevant files on compact disks and post to the person who requests the data.
Computational Techniques Used
We make use of two computational techniques - stochastic dynamics simulations and molecular dynamics calculations. Both approaches deals with many-particle systems, and their algorithms are based on, respectively, the Langevin equation and Newton's equation.
Publications, Awards and External Funding
External Funding and Awards
Our projects are supported by grants awarded to S. H. Chung by the National Health & Medical Research Council and the Australian Resarch Council.
Publications