- Visualization, Kinetics, and Thermodynamics of DNA-Protein Interactions (2005)
- In this work the two spectroscopic techniques surface plasmon resonance (SPR) and fluorescence correlation spectroscopy (FCS) as well as the imaging technique cryo-transmission electron microscopy (cryo-TEM) were used to gain kinetic, thermodynamic and structural information about DNA–protein interactions. Furthermore, the micrographs obtained by cryo-TEM were compared to AFM images taken in a previous work. The main goal of this work was to investigate the influence of surfaces on DNA–protein interactions, therefore the methods mentioned above were chosen. Both SPR and AFM deal with molecules attached to a surface, whereas FCS and cryo-TEM monitor the molecules in free solution. As a suitable model system the well characterized interaction between the human replication protein A (RPA) and DNA was chosen. The application of SPR and FCS to analysing the binding of RPA to ssDNA yields information about the kinetics and thermodynamics. No modification of the protein is required and biotinylated and fluorescently labelled DNA strands are available from commercial sources. Salt concentration, pH and temperature can be varied over a wide range. To best of our knowledge, FCS has not been used previously to obtain equilibrium constants at different temperatures. In this work it was demonstrated how temperature dependent SPR and FCS measurements can be performed and evaluated to determine thermodynamic data of DNA–protein interactions. Astonishingly, the equilibrium constant KD for the binding of RPA to ssDNA obtained by FCS is larger than the value obtained by SPR by a factor of 20–25, depending on the temperature. Therefore the values found for the Gibbs free energy were different, whereas the values for the reaction enthalpy were nearly the same for the two methods used. There are clear evidences that the difference in KD and therefore in Gibbs free energy measured by the two methods is due to different reaction entropies. In SPR the reaction is restricted to two dimensions due to immobilization of the DNA molecules to the sensor surface, thus the rate constants obtained might not be the true association and dissociation rates. As a main result, the data obtained by SPR differ from the data gained from the free solution experiments. The reason for this is a loss of one degree of freedom, which in turn results in different entropic terms for the surface and the free solution techniques. In contrast, FCS is able to follow complex formation without spatial restrictions. In consequence, the reaction in three dimensions is entropically less favourable than the reaction at the solid-liquid interface. This might be due to differences in the cratic entropy between the two geometries, however, the role of hydration can not be assessed by our experiments. The picture of the DNA–RPA interaction was completed by further FCS measurements using various dsDNA fragments containing damage sites. The binding of RPA to undamaged dsDNA fragments showed a low affinity to dsDNA (approx. 15%), as expected from previous AFM experiments. Since RPA is known to have a high affinity to singlestranded DNA, this finding may be explained by the binding of RPA to unpaired nucleotides at the end of the dsDNA. Comparing the two imaging techniques AFM and cryo-TEM one does not find a strong influence of the surface on the DNA–RPA interaction. The kinks formed by UV-damaged DNA observed in AFM experiments could not be verified by the cryo-TEM experiments. There might be two reasons for this: First, the kinks in the AFM experiments are induced by the mica surface and therefore do not occur in cryo-TEM experiments. Second, the resolution of the TEM is not as good as in AFM, therefore, the kinks can not be seen in the TEM. The question if the DNA is wrapping around the RPA as stated in earlier works can not be answered using cryo-TEM. The resolution of this method is not as good as in AFM. In order to get micrographs with a better resolution one has to perform simple TEM experiments including staining of the molecules. The drawback of this procedure is that the molecules are influenced by the staining chemicals and therefore not in their natural state. In a very last part of this work the mini-chromosome maintenance com-plex was investigated using FCS and cryo-TEM. It was shown that the protein exhibits a medium affinity to ssDNA and dsDNA. The structure of the DNA substrate does not play an important role, the interaction was the same for simple and bubble dsDNA and dsDNA containing a ssDNA tail.