Determination of electrostatic interaction energies and protonation state populations in enzyme active sites

The pH-dependence of the NMR chemical shift for titratable groups in proteins often deviate from a standard Henderson-Hasselbalch (HH) titration curve. A non-HH dependence of the chemical shift for a given residue can arise from a single-site, non-HH titrational event for that residue, or if the chemical shift of the group is influenced by additional titrational events occurring in other residues. We show that simultaneous fits of several non-HH NMR titration curves of interacting protein residues to a statistical mechanical model can be used to distinguish between these two cases. From fitting of non-HH titrations, we can extract electrostatic interaction energies between protein residues. Furthermore, by performing simultaneous fits of NMR titration curves and enzymatic pH-activity profiles, we can gain information on the identity and populations of the catalytically competent protonation states in enzymes. We apply the global fitting of titrational events (GloFTE) method to experimental data on five enzyme systems and on a single non-enzyme system, and show that the extracted electrostatic interaction energies and effective dielectric constants for a subset of these systems agree excellently with experimentally determined values as well as with theoretical calculations. In the case of reduced Escherichia coli thioredoxin we use GloFTE analysis to distinguish between two possible interpretations of the NMR titration curves of the active site residues. We also show that for the strongly coupled system of titratable groups in the active site of the Bacillus circulans xylanase (BCX) N35D mutant, GloFTE fits of a single titration curve and an enzymatic pH-activity profile can give a full description of the energetics of the titrational events in the enzyme's active site. Using only the X-ray crystallographic structure of the enzyme and the electrostatic interaction energies extracted from such a GloFTE fit, we can uniquely identify the three catalytic groups in this system. This raises the prospect of completely characterising active site titrational events from a single unassigned NMR titration curve and an enzymatic pH-activity profile.