HYDROGEN BONDING IN INHOMOGENEOUS ENVIRONMENTS
EMSL Project ID
20499
Abstract
We propose to improve the treatment of hydrogen bonds in simulations, especially of proteins. These will be incorporated into suites of programs that carry out Monte Carlo, and eventually Molecular Dynamics, simulations. Because computer simulation has become one of the primary means of determining the function of proteins, including membrane proteins like ion channels, our primary interest, the importance accurate treatment of hydrogen bonds is tremendous. Furthermore, the conformational changes that occur in the course of the functioning of proteins are also subprocesses of protein folding, one of the major unsolved problems in biophysics. Thus the problem we are setting out to solve impacts at least three of the major aims of this round of Computational Grand Challenges: since proteins function essentially as interfaces, and membrane proteins are parts of interfaces, the chemistry at interfaces question is relevant. Second, so is computational biology, given the biological applications; there is a clear multiscale aspect here, as we go from quantum treatment of individual hydrogen bonds to understanding the behavior of entire membrane proteins, from a scale of about 0.1 nm to a scale of about 100nm, or three orders of magnitude. Third, and most direct, is the biological interactions goal, as channels are parts of membrane-bound receptors, coupling signaling molecules to intramolecular responses; simulations of complete receptors will be facilitated by the existence of an accurate and efficient hydrogen bond potential. Protein folding, and even further afield, applications to hydrogen bonds in fuel cell membranes, are also likely, but not primary aims of the work proposed here. The basis for the work is our observation that hydrogen bond energy is dependent on the neighboring hydrogen bonds that two hydrogen bonded molecules form. So far, water clusters have shown the importance of cooperative effects that must be taken into account in simulations, but in practice usually are not. The energies can vary by up to 3 kBT, easily large enough to require attention in a simulation. First, the potentials must be extended to systems other than water, and relevant to protein structure and function, including hydronium ions (Eigen and Zundel) and hydroxide ions. Next it will be necessary to obtain potentials for a variety of amino acids that are capable of hydrogen bonding; this will unquestionably require the use of the supercomputer. The new potentials have be the incorporated into simulation code. The first code to be used will be the MMC program, developed by one member of this team (MM), and which continues to allow further development. This will make it possible to produce simulations of systems that are of biological interest, starting with ion channels such as MscL and MscS. Simulations already carried out on these channels, including work by two members of the team (SS,AA), have shown (at the level of potentials now in use, including TIP3P water) gating that is water dependent, with the open-closed transition a wetting-dewetting transition. Although these results are believed to be qualitatively correct, these simulations predicted a biased onset of the wetting, much earlier than follows from experimental data. This illustrates the importance of the correct potentials for water, as the structure of the open and closed states of the channel is a function of the hydrogen bonding of the water in the confines of the channel. Other cases may not work out as well. Development of the improved potentials requires the use of supercomputer facilities that are adequate to handle quantum calculations with upwards of 2000 basis functions, and that are capable of carrying out the large simulations that will be required to test the functions, and to apply them.
Project Details
Project type
Capability Research
Start Date
2006-10-01
End Date
2009-09-30
Status
Closed
Released Data Link
Team
Principal Investigator
Team Members
Related Publications
Green ME. 2008. "Consequences of Phosphate-Arginine Complexes in Voltage-Gated Ion Channels." Channels 2(6):395-400. doi:http://www.landesbioscience.com/journals/channels/
Kariev AM, and ME Green. 2008. "Quantum Mechanical Calculations on Selectivity in the KcsA Channel: The Role of the Aqueous Cavity." Journal of Physical Chemistry B 112(4):1293-1298. doi:10.1021/jp076854o
Kariev AM, VS Znamenskiy, and ME Green. 2007. "Quantum Mechanical Calculations of Charge Effects on gating the KcsA channel." Biochimica et Biophysica Acta--General Subjects 1218-1229.
Znamenskiy VS, and ME Green. 2007. "Quantum Calculations on Hydrogen Bonds in Certain Water Clusters Show Cooperative Effects." Journal of Chemical Theory and Computation 3(1):103-114. doi:10.1021/ct600139d