Kurzbeschreibung
Bovine pancreatic Ribonuclease A (RNase A) was used as a model system to examine the relationship between local protein dynamics and ligand binding specificity. The extremely well documented biochemistry and comparatively smaller size of RNase A make it an ideal model system for this study. A commercially available series of derivatized mononucleotide ligands with systematic structural perturbations were used to probe the binding specificity of the protein. The ligands were chosen to mimic the post-cleaved state of RNA, with two binding at the purine (B1) and two at the pyrimidine (B2) subsites. The global thermodynamic parameters for each binding reaction were determined using Isothermal Titration Calorimetry, and the local entropic contributions for the binding events were subsequently evaluated using high-resolution solution Nuclear Magnetic Resonance Spectroscopy. Isothermal titration calorimetric studies were performed to obtain the dissociation constants and global thermodynamic parameters for each binding event. All four reactions were spontaneous and enthalpically driven, with global DeltaG° values of -5.46+/-0.03, -6.04+/-0.04, -4.53+/-0.03 and -6.69+/-0.06 kcal/mol for the binding of AMP, GMP, CMP and TMP respectively. The two ligands binding to the purine subsite displayed a greater disparity between the global enthalpy (DeltaDeltaH°=6.2 kcal/mol) and entropy (DeltaDeltaS°=17.4 cal/mol-K) values, compared to the pyrimidine ligands (DeltaDeltaH°=0.8 kcal/mol and DeltaDeltaS°=0.5 cal/mol-K), indicating that subtle structural changes in the ligands are reflected through significant changes in binding thermodynamics. Spin-lattice and spin-spin relaxation experiments and steady-state heteronuclear Nuclear Overhauser Effect experiments were performed at 500 and 600 MHz for the free protein and each ligand-bound complex. The relaxation data were analyzed using the FAST-Modelfree in combination with Model-Free 4.02 to obtain overall and internal dynamics parameters for the backbone amide vectors for each case. The value of the average order parameter (S2) remains relatively unchanged at approximately 0.83+/-0.05, however, the active site residues show statistically significant changes in the difference between S 2 values for each binding event, indicating that the dynamics changes that occur at the active site are compensated by motions in other areas of the protein. A small change in electrostatics between the purine ligands pushes GMP to bind at the pyrimidine subsite, while the extra methyl group in TMP seems to cause a decrease in its binding constant. Further dynamics experiments on the methyl groups and side chains would provide more insight into the general relationship between conformational entropy, binding specificity and ligand structure.