Book of Abstracts: Albany 2009

category image Albany 2009
Conversation 16
June 16-20 2009
© Adenine Press (2008)

Molecular Mechanics Analysis of Minimal Energy RNA Conformational Change Pathways

Conformation changes are important in RNA for both binding and catalysis. We are developing computational methods for exploring and understanding pathways for defined conformational changes.

One system of study is the conformational change of a non-canonical pair. In an NMR structure of an AA mismatch in the sequence:

[Chen, G., et al. (2006) Biochemistry, 45(22):6889-6903] (P = purine), the AA non-canonical pair is in conformational exchange between a minor and major conformations. The conversion of the major trans Hoogsteen-sugar to the minor trans sugar-Hoogsteen non-canonical pair occurs on the NMR timescale.

We used the AMBER molecular mechanics software package to model conformational change pathways. Initial modeling was done with Targeted Molecular Dynamics (TMD), which applies a biasing potential based on RMSd to a target structure in MD simulations. This provides a forced approximation of possible pathways with dynamics. We further used Nudged Elastic Band (NEB), which predicts minimal potential energy paths using a series of all atom images of the system along the path. Both TMD and NEB provide insight into conformational change pathway. TMD provides a rough approximation of pathways undergoing dynamics in time, while NEB provides a time-independent and discrete low potential energy pathway.

Predicted pathways from NEB were analyzed and a reaction coordinate determined for the conformational change. This reaction coordinate involves an improper dihedral angle defined by C8, C4, and N1 on one adenine and C5 on the second adenine in the non-canonical pair. The minor state has an improper dihedral value of about 0 degrees, while the major state has a value of about +/- 180 degrees. Umbrella sampling was then used to predict the free energy profile along the 360 degree reaction coordinate. Umbrella sampling was done using 36 windows of 10 degrees each with 12 ns of sampling per window for 6 different random number seeds. Total sampling involved about 2.6 microseconds of MD spanning about 7 total years of CPU time. The free energy profile suggests errors in the AMBER force field because there is a reversal in the relative free energies of the major and minor structures.

Keith Van Nostrand
Scott D. Kennedy
David H. Mathews

Department of Biochemistry and Biophysics
University of Rochester Medical Center
601 Elmwood Ave. Box 712
Rochester, NY 14642

Phone: 585-275-1754
Fax: 585-275-6007
E-Mail: keith Van Nostrand