Title: Computational Alanine Scanning with Linear Scaling Semi-Empirical Quantum Mechanical Methods

Abstract: Several areas of active interest – such as cancer targets, antibodies, signaling, and so on – have been shown to be related to protein-protein interfaces. In order to study these types of interfaces, alanine scanning is a powerful experimental tool for understanding the key interactions. At the same time, linear scaling, semi-empirical quantum mechanical calculations are now sufficiently fast and robust to allow meaningful calculations of large systems such as proteins, RNA and DNA. In particular, they have proven useful in understanding protein-ligand interactions. Here we ask the question: can these linear scaling quantum mechanical methods developed for protein-ligand scoring be useful for computational alanine scanning? To answer this question, we assembled 15 protein-protein complexes with available crystal structures and sufficient alanine scanning data. In all, the data set contains DDGs for over 400 single point alanine mutations of these 15 complexes. We show that with only one adjusted parameter the quantum mechanics based methods out perform both buried accessible surface area and a potential of mean force and compare favorably to a variety of published empirical methods. In order to further understand the quantum-scoring results, the mutation effect was studied using the residue-to-residue QM-based, pairwise interaction energy decomposition (QM-PWD) method. The difference of PWD values between wild-type and mutated complexes associated with the mutated residue was designated as the primary mutation effect. As expected, in most cases this primary effect was the most significant effect of mutation. However, when considering the secondary effects – or those effects that result from the total mutation effect minus the primary effect – four out of 14 mutated complexes have higher secondary mutation effects. Further, three out of 14 mutated complexes have small total mutation effects and their primary mutation effects can be neglected. This result suggests that in order to truly understand protein-protein interactions, one must fully understand the entire surface chemistry. Taken together, we have demonstrated that QMScore along with QM-PWD can indeed be applied to complexes involving large protein/ligand interfaces.

Authors: Lance M. Westerhoff, David J. Diller, Xiaohua Zhang, Christine Humblet

Conference: Keystone Symposia – Computer-Aided Drug Design (Z7).