Computational studies of biological macromolecules are challenging due to large size of biomolecules, their conformational flexibility, and the need in explicit water solvation in order to simulate conditions close to experiment. Under these circumstances studying molecular systems via quantum-mechanical calculations becomes exceedingly difficult. Natural is the attempt to reduce the complex quantum-mechanical picture to a more tractable one by accommodating classical-mechanical principles. However, the simplified models may overlook important physics details of atomic interactions. To avoid such potential pitfalls higher level of theory methods should be available to conduct validation studies. Using semiempirical linear scaling quantum-mechanical LocalSCF method we performed molecular dynamics simulation of ubiquitin in explicit water. The simulation revealed various deviations from the classical mechanics picture. The average charge on amino acids varied depending on their environment. We observed charge transfer channels transmitting electric charge between amino acids in sync with protein motion. We also noticed that the excess charge transferred from protein to water creates a charge cloud around the protein. The observed global dynamic effects of charge transfer represent a new previously unaccounted degree of freedom of biomolecules which requires QM treatment in order to obtain more accurate dynamics of biomolecules at atomic resolution.
NDDO method PM5 Hamiltonian Ubiquitin Water droplet Spherical boundary potential QM MD Charge transfer VFL approximation LocalSCF Linear scaling
Austin model 1
Density functional theory
Hartree–Fock method using Pople 6-31G* basis set
Local self consistent field
Neglect of diatomic differential overlap
Parametric method 3
Parametric method 5
Recife model 1
Variational finite localized molecular orbital approximation
This is a preview of subscription content, log in to check access.
The Authors are grateful to the Texas Advanced Computing Center (TACC) (http://www.tacc.utexas.edu) for providing high-performance computing resources for this project. Support from a R.A. Welch Foundation Chemistry and Biology Collaborative grant from the John S. Dunn Gulf Coast Consortium for Chemical Genomics is greatly acknowledged.