GPU accelerated molecular dynamics simulations for protein-protein interaction of ankyrin complex

• Main Authors: Kanchanok Kodchakorn, Supaporn Dokmaisrijan, Wei Lim Chong, Apirak Payaka, Tanchanok Wisitponchai, Piyarat Nimmanpipug, Sharifuddin M. Zain, Noorsaadah Abd Rahman, Chatchai Tayapiwatana, Vannajan Sanghiran Lee
• Format: Journal
• Published: 2018
• Subjects: Engineering; Materials Science; Physics and Astronomy
• Online Access: https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84903162566&origin=inward; http://cmuir.cmu.ac.th/jspui/handle/6653943832/53487
• Institution: Chiang Mai University

The ankyrin repeat (AR) can be used as a versatile scaffold for protein-protein interactions. It consists of a 33-residues sequence motif found in proteins with diverse functions, such as transcription initiation, cell cycle regulation, cytoskeletal integrity, ion transport, and cell-cell signaling. Using AR with high affinity for the Escherichia coli maltose binding protein (MBP) as our model system, we explored a structure-based computational protocol to probe and characterize binding affinity hot-spots at protein-protein interfaces. In this study, the long time scale dynamics simulations with GPU accelerated molecular dynamics (MD) simulations in AMBER12 have been performed to locate the hot-spots of protein-protein interaction by the analysis of the Molecular Mechanics-Poisson-Boltzmann Surface Area/Generalized Born Solvent Area (MM-PBSA/GBSA) of the MD trajectories. The two designed AR systems with different binding affinities from ELISA were simulated. Our calculations gave the absolute binding affinity predictions which are in agreement with the kinetic experiment. The difference in binding affinity of the two selected clones is due to the framework mutations which are mostly conserved at a β-hairpin/loop region. AR domain is most probably not affected by the alteration of this framework from the long time scale MDs. © 2014 Taylor & Francis Group, LLC.