Computational studies of protein pK(a)s and metalloprotein reduction potentials

Protein pKas and metalloprotein reduction potentials are studied with computational methodologies based on an ab initio quantum mechanics (QM) description of the protein and a linearized Poisson-Boltzmann Equation (LPBE) description of the solvent. The practical applicability of the QM/LPBE method is extended to proteins by using a QM description of the ionizable residue and a molecular mechanics (MM) description of the rest of the protein. This QM/MM/LPBE method is used to predict the pKa of Lys55 in the serine protease inhibitor turkey ovomucoid third domain (OMTKY3) and the prediction of 11.0 is in good agreement with the experimental value of 11.1. This is the first time a protein pKa value has been predicted with QM/MM methods. The QM/LPBE method is used to predict and interpret the pKa’s of the five carboxyl residues (Asp7, Glu10, Glu19, Asp27, and Glu43) in OMTKY3. All the predicted pKa values are within 0.5 pH units of experiment, with a root mean square deviation of 0.31 pH units. We find that the decreased pKa’s observed for some of the residues are primarily due to hydrogen bonds to the carboxyl oxygens. Hydrophobic effects are also shown to be important in raising the pKa. Interactions with charged residues are shown to have relatively little effect on the carboxyl pKa’s in this protein, in general agreement with experiment. The relative Cu/Cu reduction potentials of six type-1 copper sites (cucumber stellacyanin, P. aeruginosa azurin, poplar plastocyanin, C. cinereus laccase, T. ferrooxidans rusticyanin and human ceruloplasmin), which lie in a reduction potential range from 260 mV to over 1000 mV, have been studied with the QM/LPBE method. For