Energetic contribution of side chain hydrogen bonding to the stability of staphylococcal nuclease

Hydrogen bonds are a ubiquitous feature of protein structures, yet there is great uncertainty about the energetic contribution of hydrogen bonding to protein stability. This study addresses this question by making a series of single substitution mutations in the model protein staphylococcal nuclease. These mutants have had a residue capable of participating in hydrogen bonding either removed or introduced. The variants we have investigated are as follows: nine valines substituted with threonine and serine; eight threonines converted to valine, serine, and cysteine; and seven tyrosines replaced by phenylalanine and leucine. The stabilities of these 56 mutant proteins were determined by titration with guanidine hydrochloride using fluorescence as a probe of structure. In general, it was found that the stability effects of removing a hydrogen bonding residue and replacing it with a nonbonding residue were relatively small. This was true even in the case of buried residues participating in hydrogen bonds, where the substituted residue leaves an unfulfilled hydrogen bond in the hydrophobic core. In contrast, introducing a hydrogen bonding residue in place of a nonbonding residue was generally more costly energetically. A wide variability in the cost of burying a hydroxyl was observed, but this does not seem to be due to differences in hydrogen bonding. The overall energetic contribution of various wild-type hydrogen bonding interactions was evaluated as being favorable. A range of energies from approximately 1.5 to 4.0 kcal/mol was estimated for the contribution of these interactions to the stability of the native state.