Influence of Conformational Averaging on 1H–H NOEs and Structure Determination in DNA

✍ Scribed by Andrew N. Lane

• Publisher: John Wiley and Sons
• Year: 1996
• Tongue: English
• Weight: 768 KB
• Volume: 34
• Category: Article
• ISSN: 0749-1581
• DOI: 10.1002/(sici)1097-458x(199612)34:13<s3::aid-omr15>3.0.co;2-q

✦ Synopsis

The influence of conformational averaging in nucleotides on NOES in DNA was investigated. DNA decamers containing both N and S-type sugar puckers were constructed and energy minimized. NOE time courses were calculated for the entire spin system of each structure and for ensembles of these conformations. Restraints on sugar conformation caused only local changes in the DNA structure, mainly readjustments of the glycosidic torsion angles and the backbone angles 5 and j 3. Restrained molecular dynamics calculations showed that local conformations are well determined using tight constraints (f0.2 8, for r < 2.6 and f0.3 8, otherwise), and less well determined using loose constraints corresponding to strong, medium and weak NOES ( < 2.6, < 3.5 and < 5 A),

respectively. In the presence of conformational averaging, tight constraints produce structures that are incorrect at the local level, whereas loose constraints produce structures that are either imprecise or inaccurate (or both).

Fitting NOE time courses provides accurate local conformational parameters that can be used to generate tight constraints for restrained molecular dynamics calculations. For conformational mixtures where the relative populations are approximately known, fitting ensemble-average NOES to the observed NOE time courses also provides accurate parameters for both structures in a mixture of two equally populated conformations, and for the major conformer when the populations are strongly biased. Structures generated from molecular dynamics calculations with several restraint sets corresponding to different nucleotide conformations did not account for the ensembleaveraged NOEs. The ensemble of these structures, however, gave reasonable agreement with the input NOE data.