We have focused on DNA complexes of restriction endonucleases in particular. Beyond their unparalleled importance as tools for the analysis and manipulation of DNA, restriction enzymes have proven remarkable model systems for studying numerous aspects of protein-nucleic acid interactions. These proteins combine high binding strength and extraordinary sequence specificity. We take advantage of the unique experimental tools we have developed to investigate the coupling of structure, thermodynamics and function of these complexes from a fresh perspective. We are currently investigating DNA complexes of the type II restriction enzyme, EcoRV. Typically restriction endonucleases can distinguish between specific recognition and nonspecific DNA sequences quite efficiently in the absence of divalent metal co-factors. At present, however, results in literature suggest that EcoRV has a quite low sequence stringency. The majority of studies show only weak selectivity, below 10 for the relative specific-nonspecific competitive binding constant. Only one group has reported value of 120 under the same experimental conditions. It has been suggested that EcoRV represents a new paradigm for restriction nuclease recognition. X-ray crystal structures are available for both cognate and non-cognate EcoRV complexes in the absence and in the presence of divalent ions. Because the two structures are so substantially different, it is contra intuitive that specific and nonspecific binding should barely differ in binding energy in the absence of divalent ions. The interface of the specific complex is essentially anhydrous with many direct DNA-protein interactions and is very different from the non-cognate complex that has a large water filled gap at the protein-DNA interface. The novel self-cleavage assay we developed recently is broadly applicable to measuring DNA-protein interactions, particularly DNA binding of restriction endonucleases. This solution technique uses the cleavage reaction of restriction endonucleases to measure sensitively their binding to DNA. We used the self-cleavage assay to quantitate EcoRV-DNA binding. Kinetic studies are necessary to ensure that equilibrium studies can be performed properly. We found that EcoRV association kinetics is very complicated. Surprisingly, the association kinetics of EcoRV shows at least two components;one that is quite fast as is typical for specific binding. The other is unusually slow with a half-life time that can range from about 20 min. at pH 6.9 up to an hour at pH 7.6. This is far different from our previous observations with EcoRI. The slow component of the association kinetics indicates that at least 2 hours incubation pH 6.9 and even longer time at pH 7.6 is necessary to reach equilibrium. We consider the unusual association kinetics of EcoRV the most promising of our findings for explaining the significant variation in the specific equilibrium constant values reported by different groups. It would be easy to underestimate amount of bound protein (and consequently binding constant) if the incubation time was not long enough to reach equilibrium. We found that EcoRV can effectively distinguish between cognate and nonspecific DNA sequences in the absence of divalent co-factors though, as was shown before, divalent co-factors dramatically increase EcoRV binding selectivity. We demonstrated that both pH and osmotic stress are also critically important for the ability of the enzyme to bind DNA in a specific manner. Ratio between specific and nonspecific binding constants increases from 56 at pH 7.6, to 280 at pH 6.9, and to 1085 at pH 6.2 in the absence of osmolytes. Even at pH 7.6, the ratio between specific and nonspecific binding constants increases from 56 in the absence of neutral solutes to 3300 in the presence of 1 osmolal triethylene glycol, mimicking the crowded environment of the living cell. EcoRV should not be considered a representative of a different class of restriction enzymes that is capable of sequence recognition at the cleavage step of the reaction only. We found that the free energy difference between specific and nonspecific EcoRV complexes is linearly dependent on solute osmolal concentrations for each of the four solutes used. We can determine the difference in sequestered water between the complexes through the dependence of the relative specific-nonspecific binding constant on solution osmotic pressure. We found that the difference in hydration between specific and nonspecific EcoRV complexes depends on solute nature changing from 116 water molecules measured for betaine up to 240 water molecules measured for the triethylene glycol. This result suggests significant difference in surface exposed area between two complexes. Manuscript is in preparation for this part of the project. We have also further developed a method for stabilizing labile DNA-protein complexes for analysis by the gel mobility shift assay. The electrophoretic mobility shift assay (EMSA) is a standard and widely used tool in molecular biology for measuring DNA-protein complex formation. Many nonspecific DNA-protein complexes, however, are weak enough that they dissociate in the gel, giving smeared bands that are difficult to quantitate precisely. In order to extend the applicability of the EMSA to these labile complexes, we have investigated the effect on adding stabilizing osmolytes (such as glycerol or triethylene glycol) to the gel itself. This project is a logical continuation of our previous work on trapping DNA-protein complexes. The dissociation of complex in a gel necessarily exposes the protein and DNA surface area that was buried in the complex. The stabilizing effect of osmolytes on protein-DNA complexes can be rationalized by the exclusion of osmolytes from exposed protein and DNA surfaces. In this work, we have focused on complexes of the restriction endonuclease EcoRI with nonspecific and star sequence oligonucleotide. Our results clearly demonstrate that including the osmolyte triethylene glycol in the gel dramatically stabilizes both the weak star and nonspecific complexes of EcoRI. Without solute, both non-cognate and nonspecific complexes simply dissociate too quickly in 10% polyacrylamide gels. We showed that 30% triethylene glycol in the gel (equivalent to 4.3 osmolal) is enough to stabilize completely complexes that have dissociation constants at regular salt and pH conditions in the micromolar range. The technique can be readily used for even weaker complexes and in principle for any RNA and DNA-protein complex that is sensitive to osmotic stress. Extension of this approach to other techniques for separating complex and free components as gel chromatography and capillary electrophoresis is straightforward.
| Sidorova, Nina Y; Scott, Thomas; Rau, Donald C (2013) DNA concentration-dependent dissociation of EcoRI: direct transfer or reaction during hopping. Biophys J 104:1296-303 |
| Sidorova, Nina Y; Muradymov, Shakir; Rau, Donald C (2011) Solution parameters modulating DNA binding specificity of the restriction endonuclease EcoRV. FEBS J 278:2713-27 |
| Sidorova, Nina Y; Hung, Stevephen; Rau, Donald C (2010) Stabilizing labile DNA-protein complexes in polyacrylamide gels. Electrophoresis 31:648-53 |
| Rau, Donald C; Sidorova, Nina Y (2010) Diffusion of the restriction nuclease EcoRI along DNA. J Mol Biol 395:408-16 |
