|Speaker:||Craig J. Benham|
Mount Sinai School of Medicine
Stress-Induced DNA Duplex Destabilization:|
|Date:||Tuesday, November 17, 1998|
|Place:||Gould-Simpson, Room 701|
DNA within living organisms is topologically constrained in a way that imposes stresses on the molecule. These stresses can destabilize the B-form duplex, causing local strand separations to occur at the sites where the thermodynamic stability is least. Computational methods have been developed to predict the locations and extents of destabilization in stressed DNA sequences. Although these analyzes have no free parameters, their results agree precisely with experimental determinations of the extents and locations of denatured regions. This allows their use to predict the destabilization properties of other sequences, on which experiments have not been performed.
When genomic DNA sequences are analyzed in this way, the sites of predicted duplex destabilizations are found not to occur at random, but instead are closely associated with several specific types of DNA regulatory regions. In several cases these regulatory regions have no consensus sequence or motif, so the determinants of their positional specificity were not previously known. The most strongly destabilized sites occur in the 3' flanks of genes. This pattern is found in genes from bacteria, viruses, yeast and higher organisms. Origins of replication also contain regions that are susceptible to stress- induced DNA strand separation. The third class of regulatory regions that exhibit characteristic destabilization properties are sites of DNA attachment to the nuclear scaffold, whose activities regulate chromosomal architecture. Here the extent of destabilization correlates closely with the strength of scaffold attachment. Experiments confirming these theoretical predictions have been performed in several cases.
This talk will present a brief description of the biological background to this problem. Some of the techniques used to analyze duplex destabilization in topologically stressed domains will be sketched. A selection of predictions regarding the destabilization properties of regulatory regions will be presented, experiments testing these predictions will be described, and the implications concerning possible regulatory mechanisms will be considered. The incorporation of these methods into new computational strategies to search genomic sequences for regulatory regions will be discussed.