Modeling DNA Thermodynamics under Torsional Stress

Negatively twisted DNA is essential to many biological functions. Due to torsional stress, duplex DNA can have local, sequence-dependent structural defects. In this work, a thermodynamic model of DNA was built to qualitatively predict the local sequence-dependent mechanical instabilities under torsi...

Full description

Saved in:
Bibliographic Details
Published in:Biophysical journal 2014-03, Vol.106 (5), p.1182-1193
Main Authors: Wang, Qian, Pettitt, B. Montgomery
Format: Article
Language:English
Subjects:
Base Pairing
Base Sequence
Biophysics
Deoxyribonucleic acid
DNA
DNA - chemistry
DNA - genetics
Energy
Models, Molecular
Osmolar Concentration
Proteins and Nucleic Acids
Rotation
Shear stress
Simulation
Stress, Mechanical
Temperature
Thermodynamics
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Negatively twisted DNA is essential to many biological functions. Due to torsional stress, duplex DNA can have local, sequence-dependent structural defects. In this work, a thermodynamic model of DNA was built to qualitatively predict the local sequence-dependent mechanical instabilities under torsional stress. The results were compared to both simulation of a coarse-grained model and experiment results. By using the Kirkwood superposition approximation, we built an analytical model to represent the free energy difference ΔW of a hydrogen-bonded basepair between the B-form helical state and the basepair opened (or locally melted) state, within a given sequence under torsional stress. We showed that ΔW can be well approximated by two-body interactions with its nearest-sequence-neighbor basepairs plus a free energy correction due to long-range correlations. This model is capable of rapidly predicting the position and thermodynamics of local defects in a given sequence. The result qualitatively matches with an in vitro experiment for a long DNA sequence (>4000 basepairs). The 12 parameters used in this model can be further quantitatively refined when more experimental data are available.
ISSN:0006-3495
1542-0086
DOI:10.1016/j.bpj.2014.01.022