Introduction

“SIMPLE” is a web tool that estimates the free energy change due to an aliphatic mutation of proteins. SIMPLE calculates the change in the van der Waals contribution, hydrophobicity, electrostatics interactions and configurational entropy between the folded and unfolded state of aliphatic point mutations based on an empirical free energy scoring function. Users submit the protein structure file and indicate up to 10 single residue mutations. The output will be accessed online and emailed back to the users, including total free energy change and each of the energetic terms used. SIMPLE has been successfully tested and validated by a 182 global aliphatic mutation dataset and a 94 alanine mutation subset, predicting 20%–30% less false positives and yielding more accurate predictions than any published empirical energy function. The simplicity of the SIMPLE energy function allows a physical interpretation of the contribution of the different energy terms to the change in stability, offering a unique biophysics insight to predict the protein stability of a mutant protein.

Content:

• Method
  
• Results
• Comparison with other methods
  
• Publication and reference
  

Method

Energy Scoring Function

The change in free energy due to a point mutation of an aliphatic residue is estimated based on a based on an empirical free energy scoring function. When a wild-type i residue is mutated to a j residue, the change in free energy difference of folding is related through the following thermodynamic identity to the changes in free energies that take place in the folded and unfolded states:

A general way to partition the Gibbs free energy difference (assuming fixed backbone) between the mutant and wild-type proteins is the following:

Computational Method

The computation approach followed is summarized in following figures and consists on the following steps:

• Preprocessing of wild-type structures. All ions and water molecules were stripped off the PDB file and the protein structures were minimized by running adopted basis Newton–Raphson (ABNR) minimization steps in CHARMm27b3 (using version 19 parameters)
  with fixed backbone coordinates.
  
• Preprocessing of mutant structures. For strict side chain shortening mutations the mutant residue atoms retained the coordinates of the modeled wild-type protein. Otherwise the side chain of the mutated residue was created using SCWRL 3.0.
  
• Calculation of energy terms. Free energy terms were calculated on the minimized structures using the CHARMm force-field. Van der
  Waals (vdW) energy and electrostatic free energy terms were calculated using the ACE force-field with the standard parameters. Then, energy change of vdW and elec were calculated as the difference in these energy terms between mutant and wild-type.
  

Fig. 1: Flowchart for the computational approach used by SIMPLE. First, the structure of the wild type protein is needed. The mutant protein is modeled using a rotamer predictor, SCWRL. Both wild type and mutant structures are minimized. Then, folded non-covalent energy terms were calculated using CHARMM with an implicit solvent model force field and the accessible surface area using NACCESS. All the unfolded state energy terms came from tabulated values extracted from available theoretical estimations in alanine tripeptides. The differences in the electrostatic energies calculated for the denatured state are computed using the CHARMm force-field and ACE on Ala-Xxx-Ala tripeptides.

Fig.2: Sketch of the folded and unfolded state of a protein and its mutant. Indicated are the free energy terms used to estimate the change of affinity when a residue i is mutated to a residue i. We note that the van der Waals interaction is the only term that cannot be estimated based on first principles. However, we argue that, for the most part, van der Waals in the folded state is compensated by interactions in the unfolded state. As suggested by the figure, electrostatic interactions for aliphatic residues in the protein surface (not buried) are effectively compensated between the folded and unfolded state.

Publication and reference

• Bueno M, Camacho CJ, Sancho J (2007) SIMPLE estimate of the free energy change due to aliphatic mutations: superior predictions based on first principles. Proteins 68: 850-862.