Dynamics in DNA damage detection and repair

DNA repair systems face a seemingly impossible dilemma; they must be extremely specific to avoid 'repairing' normal DNA but must be general enough to recognize the infinite forms that damage can take. These conflicting requirements have spawned the evolution of multiple repair pathways that collectively satisfy these goals. We use NMR spectroscopy to measure the dynamic properties of repair enzymes and DNA in order to understand how damage is found via two distinct mechanisms of DNA damage recognition. Uracil base excision repair exemplifies a highly specific repair process, relying upon a specialized enzyme, uracil DNA glycosylase (UNG), to form the exclusive chemical contacts that constitute recognition of the damaged base uracil. Relaxation dispersion dynamics experiments on UNG as it searched for DNA damage identified a single region surrounding the enzyme's DNA interaction surface that collectively experience a 900 s-1 dynamic motion. This timescale of motion is consistent with previous kinetic observations of the enzyme moving about DNA, and together with normal mode analysis, indicates the enzyme is clamping down on DNA to form the specific contacts necessary to recognize uracil. In addition to specific repair enzymes, there are also general repair pathways that are able to recognize DNA crosslinks with disparate chemical structures by the effect they have on the properties of the surrounding DNA. Using a unique crosslink that had no effect on the secondary structure of DNA, we looked for changes in base pair opening rates that might alert repair enzymes to its presence. We found that this type of crosslink is less dynamic than both normal DNA and distorting forms of damage. Thus damage identification likely depends upon recognition of the altered topology of the covalently linked DNA. Future studies will require additional tools better matched to the dynamic properties of DNA. We report the development of a new NMR experiment utilizing what we have termed frequency labeled exchange, or FLEX, to detect exchanging protons with an increase in sensitivity of several hundred fold as compared to other types of spectroscopy. We believe this methodology will open new avenues in biological NMR as well as magnetic resonance imaging.