To determine three-dimensional conformations of DNA damaged by environmental chemical carcinogens, effective molecular mechanics search techniques have been developed to deal with the large system sizes and computational demands. First, extensive surveys of the potential energy surface are carried out by energy minimization. These search strategies rely on (1) using the reduced variable domain of torsion-angle (rather than Cartesian) space, (2) building larger units (about 12 base pairs) on the basis of structures of small modified subunits, and (3) employing penalty functions to search for selected hydrogen bonding patterns and to incorporate interproton distance bounds when available from experimental high-resolution nuclear magnetic resonance (NMR) studies. Second, molecular dynamics simulations with solvent can subsequently be employed to probe conformational features in the presence of polymerase enzyme responsible for DNA replication, using structures computed in the energy minimization searches as initial coordinates. A key structure-function relationship involving mirror-image molecules with very differing experimentally determined tumorigenic potencies has been deduced: the members of the pairs align oppositely when bound to DNA, making it likely that their treatment by replication and repair enzymes differ. This opposite orientation phenomenon, first predicted computationally (Singhet al., 1991), has been observed in experimental high-resolution NMR studies combined with our molecular mechanics computations in a number of different examples and has recently been confirmed experimentally in other laboratories as well (reviewed in Geacintovet al., 1997). Elucidation of this conformational feature has paved the way to uncovering the structural origin underlying very different biological outcomes stemming from chemically identical but mirror-image molecules. © 1999 Academic Press.
CITATION STYLE
Broyde, S., & Hingerty, B. E. (1999). Effective Computational Strategies for Determining Structures of Carcinogen-Damaged DNA. Journal of Computational Physics, 151(1), 313–332. https://doi.org/10.1006/jcph.1998.6172
Mendeley helps you to discover research relevant for your work.