The low barrier hydrogen bond (LBHB) proposal revisited: the case of the Asp... His pair in serine proteases.

The fact that hydrogen bonds (HBs) can provide major stabilization to transition states (TSs) of enzymatic reactions is well known. However, the nature of HB stabilization has been the subject of a significant controversy. It is not entirely clear if this stabilization is associated with electrostatic effects of preorganized dipoles or with delocalized resonance effects of the so-called low barrier hydrogen bond (LBHB). One of the best test cases for the LBHB proposal is the complex of chymotrypsin and trifluoromethyl ketone (TFK). It has been argued that the pK(a) shift in this system provides an experimental evidence for the LBHB proposal. However, this argument could not be resolved by experimental studies. Here we explore the nature of the Asp102-His57 pair in the chymotrypsin-TFK complex by a systematic computational and conceptual study. We start by defining the LBHB proposal in a unique energy-based way. We show that a consistent analysis must involve a description in terms of the energy of the two resonance structures and their mixing. It is clarified that LBHBs cannot be defined according to strength or distance, because ionic HBs can also be strong and short. Similarly, NMR chemical shifts and fractionation factors cannot be used to identify LBHBs in a conclusive way. It is also clarified that HBs with a significant asymmetry cannot be classified as LBHBs, because this contradicts the assumption of equal pK(a) of the donor and acceptor. Thus, the main issue is the DeltapK(a) and the corresponding energy difference. With this definition in mind, we calculate the free energy surface of proton transfer in this pair and evaluate the energetics of the different ionization states of this system. The calculations are done by both the semimacroscopic version of the protein dipoles Langevin dipoles (PDLD/S-LRA) model and by the empirical valence bond (EVB) method. The calculations establish that the LBHB proposal is not valid in the chymotrypsin-TFK complex and in other serine proteases. Although previous theoretical studies reached similar conclusion, this is the first time that the same set of free energy calculations reproduce all the known pK(a) values and pK(a) changes in the system, while evaluating the energetics and covalent character of the His-Asp system. The present study provides a support to the idea that enzymes work by creating a preorganized polar environment. Copyright 2004 Wiley-Liss, Inc.