Molecular Dynamics Simulations Of Biological Macromolecules

National Heart, Lung, And Blood Institute

Investigators

Linked publications, trials & patents

Abstract

Markov state model for protein conformational fluctuation: Misfolding of -2 macroglobulin (2m) into an aggregation prone intermediate state is believed to be the first stage of its aggregation and amyloid formation. High exposure of hydrophobic residues to the solvent is an intrinsic property of these intermediate states. Furthermore, a variant of b2m missing the first 6 residues on the N-terminal region (N6) is proved to be highly amyloidogenic. It has been suggested that this variant has a high equilibrium population of state that resemble the intermediate state for aggregation. We are investigating thermodynamics and kinetics of conversion between the native state and the aggregation prone intermediate state of 2m and the N6 mutant using molecular dynamics simulation and Markov state model (MSM). To sample different misfolded states of 2m and N6 we first performed metadynamics simulations and clustered the simulations into 400 clusters. These were then used to start the conventional MD simulations each for 500ns. The aggregate of 200 s were used to build the Markov model. The wild type protein has a smaller misfolding time than the mutant with smaller exposure of hydrophobic surfaces which explains its lower amyloid propensity. pH dependance of a Na channel Sodium channels play an important role in electrical signaling in cells; as such they are the targets of many drugs, as well as naturally occurring toxins from plants and animals. Inhibition and/or improper functioning of sodium channels due to mutation can lead to disease. In bacterial voltage gated sodium channels, the passage of sodium ions through the pore is controlled by a selectivity filter (SF) comprised of four glutamate residues. The number of ions bound in the channel can vary, but is about 2 on average. We study the bacterial channel NavMs. Previously, we have shown with MD simulations at constant pH and with free energy perturbation that the pKa values of the four SF glutamate residues depend on the number of ions bound in the channel, and that the fully deprotonated, singly protonated and doubly protonated states of the SF are all possible at physiological pH. Based on the MD simulations of the fully open channel, we have further shown that the conductance of the channel decreases with each proton bound to the SF. Thus the conductance of the channel is pH dependent, and decreases with lowering of pH, in agreement with experiments on similar channels. We also show that the conductance depends on the lipid composition of the membrane. Currently we are investigating the mechanism of selectivity of this sodium channel. Selectivity has been shown to depend on pH, and we are investigating the selectivity for different protonation states. Quantum Mechanical Approaches for the Evaluation of the Partition Coefficients and Acid Dissociation Constants in the SAMPL7 Physical Property Challenge The ability to make accurate predictions of chemical properties is a crucial component in drug design. However, the predictive power of many computational methods is uncertain since they have not been tested on a large scale. The SAMPL Physical Property Challenge provides an excellent opportunity to test computational methods for calculating partition coefficients and acid dissociation constants, quantities pivotal to the therapeutic efficacy of drugs. In these challenges, the experimental data are withheld until the submission deadline, so the predictive value of the unbiased theoretical methods can be revealed. In our most recent participation in the SAMPL7 competition, we submitted predictions based on quantum mechanical and machine learning methods. All our methods performed consistently well. Specifically, in the partition coefficient competition, one of our predictions was ranked as one of the best five consistently well-performed methods based on various statistical metrics. Identify the conformational states of glycine receptor alpha3 through umbrella sampling Ligand-gated ion channels allows cells to respond rapidly to changes in their external environment. The structure change from one state to other states is the key to understand how ion channels function. In many cases due to the limitation in experiment, only some state structures, a closed, open state, or desensitized state, are available. A reliable method is needed to derive the structures of the missing state structures. Even in cases structures of all states are available, it is desired to understand how conformation changes during state transition. This work presents a method that utilizes the umbrella sampling to drive conformation changes from one state to the other state. A reaction coordinate describing relative orientation of lining substructures of ion channels is proposed and the free energy profile along the reaction coordinates is produced. For glycine receptor alpha-3 pentamer, we find that there are two free energy wells separated by a barrier in the free energy profile. The desensitized state corresponding to one well at a larger reaction coordinate and the closed state corresponding to the other well at a smaller reaction coordinate. The open state locates at the barrier region and the high free energy of the open state make it unstable, which is a main reason that the open state is difficult to be captured in experiment. This free energy profile also explains the observation that the open state structure quickly decays to the other states in some computational studies. Examining the conformations of these state shows that glycine binding produces an expansion movement at the ECD and TMD interface, which opens the ion gate at the middle of the ion channel. Continued opening will result in the structure decaying into the free energy well of the desensitized state. At the desensitized state, the proline residues at bottom of the ion channel close to lock up the channel. The result agrees with the close, open, and desensitized state structures of the glycine receptor alpha1. The free energy profile along the reaction coordinate provides a thermodynamic understanding of the ion channel functional states. Glycinergic cannabinoids promote analgesia via a cholesterol-dependent mechanism Accumulating evidence has suggested that cholesterol, the main component of cell membrane, can directly interact with different neurotransmitter receptors in the brain. However, relatively less is known about the molecular detail and in-vivo consequence of cholesterol-receptor interaction. Here, we test if there is a cholesterol-dependent mechanism that underlies cannabinoid suppression of chronic pain by targeting glycine receptors (GlyR). The GlyRs are closely associated with cholesterol/caveolin-rich domains at subcellular levels. Depletion of membrane cholesterol by MCD pretreatment significantly inhibits cannabinoid potentiation of IGly in spinal neurons and in HEK 293 cells expressing GlyRs. Such inhibition is fully rescued by cholesterol supplement in a concentration dependent manner. This cholesterol-dependent mechanism appears to be receptor and cannabinoid specific. Combining molecular dynamic stimulation with functional mutagenesis analysis of human GlyRs, we propose that cholesterol sits in an inter-subunit pocket and interacts with specific residue of GlyR, thereby reducing the free energy barrier of channel opening. Thus, our findings provide a lipid-raft-dependent mechanism for glycinergic cannabinoid-induced analgesic effect.