Switchability of the helical structure of DNA

Mickey Schurr
Department of Chemistry, University of Washington

DNAs in solution are NOT simply flexible "one-dimensional" crystals! Over at least parts of their sequences, native DNAs in solution apparently exhibit a cooperative equilibrium among different helical structures (or conformations) within the B-family. These alternative conformations correspond to different "basins" of the potential-of-mean-force (as a function of the relevant structural coordinates). When two slowly interconverting conformations exhibit different intrinsic curvatures, then a bend can occur either by flexing each part of the sequence at its fixed conformation, or by converting part of the sequence from one unflexed helical conformation to the other, or by some combination of both processes. However, the time required to interconvert different helical structures substantially exceeds that to relax simple flexures. Consequently, the bending rigidity should be time-dependent, being stiffer on short time-scales than at very long times. Experimental results demonstrating this phenomenon will be discussed. Evidence that sufficient coherent bending strain shifts the structural equilibrium toward an alternative, torsionally stiffer, state will also be discussed.
Due to the cooperative nature of the structural equilibrium, the average domain sizes of regions of different helical structure are rather large. Hence, local perturbations of the helical structure can influence the average structure of the flanking DNA up to rather large distances from the site of the perturbation. Evidence will be presented that various perturbations, including (i) a particular change in sequence over a small region, (ii) a change in helical structure of that sequence, and (iii) site-specific binding of a transcriptional enhancer protein, can affect the average secondary structure of the flanking DNA up to a few hundred base-pairs away from the site of the perturbation.
Evidence is presented to demonstrate the ultra-slow kinetics of these cooperative equilibria, and the ability of the enzyme, topoisomerase I, to equilibrate metastable helical structures that commonly prevail after relaxation of superhelical stress. This action is proposed to be a second function of topoisomerase I, which functions primarily to relax superhelical stress in closed circular DNAs.