Evolutionary Analysis As a Powerful Complement to Energy Calculations for Protein Stabilization

Stability is one
of the most important characteristics of proteins
employed as biocatalysts, biotherapeutics, and biomaterials, and the
role of computational approaches in modifying protein stability is
rapidly expanding. We have recently identified stabilizing mutations
in haloalkane dehalogenase DhaA using phylogenetic analysis but were
not able to reproduce the effects of these mutations using force-field
calculations. Here we tested four different hypotheses to explain
the molecular basis of stabilization using structural, biochemical,
biophysical, and computational analyses. We demonstrate that stabilization
of DhaA by the mutations identified using the phylogenetic analysis
is driven by both entropy and enthalpy contributions, in contrast
to primarily enthalpy-driven stabilization by mutations designed by
the force-field calculations. Comprehensive bioinformatics analysis
revealed that more than half (53%) of 1 099 evolution-based
stabilizing mutations would be evaluated as destabilizing by force-field
calculations. Thermodynamic integration considers both folded and
unfolded states and can describe the entropic component of stabilization,
yet it is not suitable for predictive purposes due to its high computational
demands. Altogether, our results strongly suggest that energetic calculations
should be complemented by a phylogenetic analysis in protein-stabilization
endeavors.