Session: Research Posters

Paper Number: 112240

112240 - Accelerated Molecular Dynamics Simulation for Large Conformational Changes in Proteins 

Proteins play an essential role in various biological processes, including metabolite transport, chemical concentration regulation, and catalysis. These processes rely on conformational changes, such as side-chain isomerization, domain motions, and allosteric regulations. To understand these changes and the underlying biological phenomena, computational investigations have become increasingly important.

Molecular dynamics (MD) simulation is a commonly used tool that provides a time-dependent evolution of a system by numerically calculating Newton's equation of motion. However, despite advances in computer power, MD simulations using an all-atom model and realistic potential energy remain computationally expensive. This leads to limited simulation times on the nano- and micro-second timescales. Additionally, biological phenomena of interest, such as protein folding, large conformational change, and protein-ligand interaction, occur as rare events over a long timescale, making them difficult to observe with traditional MD simulations.

To address this issue, various hyperdynamics-based techniques have been developed, including the bond-boost method (BBM), which constructs the bias potential using the change in bond lengths. However, the structural complexity of proteins has hindered the applicability of BBM to biomolecular systems. Hence, a modified BBM is suggested for investigating the conformational pathway of proteins. We use dihedral angles and hydrogen bonds that affect the conformational transition and optimize key parameters, such as the maximum bias potential and threshold, according to the characteristics of proteins.

To validate this method, we perform both conventional and accelerated MD simulations for adenylate kinase (AdK), a monomeric phosphotransferase enzyme that undergoes a large conformational change of domains during the catalytic cycle. AdK plays a crucial role in regulating cellular energy homeostasis within the cell, and alterations in its activity can lead to diseases such as hemolytic anemia, metallic disorders, and cancer. Various structures of AdK have been revealed by experimental studies, and the open and closed structures of the apo (substrate-free) and substrate binding forms, respectively, are commonly used in computational studies. AdK's structure consists of three domains: an ATP-binding (LID) domain, a CORE domain, and an AMP-binding (NMP) domain. The LID and NMP domains are dynamic structures relative to the CORE domain and play a key role in binding substrates through a large conformational change over a long timescale.

The proposed accelerated MD method investigates the closed-to-open transition from the perspective of the conformational pathway for domain movements. As a result, it successfully detected the transition pathway of AdK by exploring a wider conformational space than conventional MD, which had limitations in analyzing phenomena occurring at a long timescale.

Presenting Author: Moon Ki Kim Sungkyunkwan University

Presenting Author Biography: Prof. Moon Ki Kim received B.S. and M.S. degrees in Mechanical Engineering from Seoul National University in 1997 and 1999, respectively, and Ph.D. degree from Johns Hopkins University in 2004. He had been an Assistant Professor in the Department of Mechanical and Industrial Engineering at University of Massachusetts, Amherst from 2004 to 2008. In 2008, he joined Sungkyunkwan University, where he is currently a Professor in School of Mechanical Engineering. His research interests are focused on computational structural biology based on robot kinematics, bioinstrumentations, and multiscale modeling and simulation.

Authors:

Soon Woo Park Sungkyunkwan University
Woo Kyun Kim University of Cincinnati
Moon Ki Kim Sungkyunkwan University