Temperature Dependence of the Backbone Dynamics of Ribonuclease A in the Ground State and Bound to the Inhibitor, 5'-phosphothymidine (3'-5')-pyrophosphate adenosine 3'-phosphate
The interaction of the dinucleotide inhibitor 5‘-phosphothymidine(3‘,5‘)pyrophosphate adenosine 3‘-phosphate (pTppAp) with bovine pancreatic ribonuclease A (RNase A) was characterized by calorimetry and solution NMR spectroscopy. Calorimetric data show that binding of pTppAp to RNase A is exothermic (ΔH = −60.1 ± 4.1 kJ/mol) with a dissociation constant of 16 nM at 298 K. At this temperature, the binding results in an entropy loss (TΔS = −16.8 ± 7.3 kJ/mol) that is more favorable than that with the product analogue, 2‘-CMP (TΔS = −31.3 ± 0.9 kJ/mol). Temperature-dependent calorimetric experiments give a ΔCp for ligand binding of −230 ± 100 J/mol K. Binding of pTppAp results in noticeable effects on the backbone amide chemical shifts and dynamics. Amide backbone 15N NMR spin-relaxation studies were performed on both apo RNase A and RNase A/pTppAp as a function of temperature. At each temperature, the model-free-determined order parameters, S2, were significantly higher for RNase A/pTppAp than for the apo enzyme indicating a decrease in the conformational entropy of the protein upon ligand binding. Furthermore, the magnitude of this difference varies along the amino acid sequence specifically locating the entropic changes. The temperature dependence of S2 at each residue enabled assessment of the local heat capacity changes (ΔCp) from ligand binding. In an overall, average sense, ΔCp for the protein backbone, determined from the NMR dynamics measurements, did not differ between apo RNase A and RNase A/pTppAp indicating that backbone dynamics contribute little to ΔCp for protein−ligand interactions in this system. However, residue-by-residue comparison of the temperature-dependent change in entropy (ΔSB) between free and bound forms reveals nonzero contributions to ΔCp at individual sites. The balance of positive and negative changes reveals a redistribution of energetics upon binding. Furthermore, experiment and semiempirical estimates suggest that a large negative ΔCp should accompany binding of pTppAp, and we conclude that this contribution must arise from factors other than amide backbone dynamics.