Free energy perturbation techniques applied to subtilisin BPN' stability.

The inability to predict the effect of specific mutations on protein stability has been an area of concern to researchers in the field of protein engineering. Small stabilization free energies (5 to 15 kcal/mol) distinguish the native and the denatured states of a protein, making the rational design of protein stability a difficult challenge. Free Energy Perturbation Technique (FEPT) appears to be a method that will be important for protein engineering to meet this challenge. Not only is it a method to evaluate potential sites for mutation prior to synthesis, it identifies important atomic contributions that are responsible for the free energy changes of interest. Accuracy and speed are the principal limitations of the technique, but the powerful combination of structure, energy, dynamics and solvent make the investment of time and effort very attractive. Our examples illustrate both the power and limitations of FEPT. Using the program CHARMm, FEPT has been applied to the well known stabilizing mutations--asparagine to Serine--at residue 218 in subtilisin BPN'. In analyzing the atomic contributions that result in the increase in stability, two mutations at residue 203 were chosen to test the predictive power of FEPT. Sometimes extraordinary measures must be undertaken to sample sufficient conformational space to achieve accurate FEPT results. However we believe the method will be invaluable in the development of rules for designing a more stable subtilisin BPN'.