Protein Stability Changes upon Point Mutations Identified with a Gaussian Network Model Simulating Protein Unfolding Behavior

Abstract

Understanding the effects of missense mutations on protein stability is a widely acknowledged significant biological problem. Genomic missense mutations may alter one or more amino acids leading to increased or decreased stability of the encoded proteins. In this study, we propose a novel approach - Protein Stability Prediction with a Gaussian Network Model (PSP-GNM) to study the effect of single amino acid substitutions on protein stability. Specifically, PSP-GNM employs a coarse-grained Gaussian Network Model (GNM) that has interactions between amino acids weighted by the Miyazawa-Jernigan (MJ) statistical potential. We use PSP-GNM to simulate partial unfolding of the wildtype and mutant structures and then, use the difference in energies of the unfolded wildtype and mutant protein structures to estimate the experimentally obtained unfolding free energy change ({Delta}{Delta}G). We verify the extent of correspondence between the {Delta}{Delta}G calculated by PSP-GNM and the {Delta}{Delta}G obtained experimentally using three datasets: 350 forward mutations from 66 proteins, 2298 forward mutations from 126 proteins and 611 forward and reverse mutations from 66 proteins and observe Pearson correlation coefficient (PCC) as high as 0.58 and root mean-squared error (RMSE) as low as 1.24 kcal/mol. The performance is comparable to the existing state of the art methods. Importantly, we do observe an increase in the correlation to 0.73 and decrease in RMSE to 1.07 when considering only those measurements made close to 25{degrees}C and neutral pH, suggesting a strong dependence on temperature and pH. PSP-GNM is written in Python and is available as a free downloadable package at https://github.com/sambitmishra0628/PSP-GNM. Author SummaryUnderstanding how genomic missense mutations impact the thermodynamic stability of encoded proteins is important to understand disease etiology. Specifically, mutant proteins are often functionally inactive and underlie numerous genetic and neurodegenerative diseases. A classic example is sickle cell anemia - a single amino acid change significantly affects hemoglobins binding affinity for oxygen. To be able to identify mutations that would likely impact the protein function and stability is therefore essential and is the focus of our study. We present an approach that relies on utilizing the intrinsic dynamics of protein structures to predict the effect of single amino acid mutations (point mutations) on protein stability. In our approach, we model proteins as coarse-grained beads (amino acids) and springs (interactions), simulate protein unfolding and identify putative residue-residue contacts that are broken during the unfolding process. We demonstrate that the knowledge of broken contacts and their order is essential in describing the thermodynamic differences between wildtype and mutant proteins. We also highlight the importance of residue-residue interactions at the mutation site in the context of protein stability prediction. Our findings present novel avenues to interpret how genomic mutations may manifest in the encoded proteins.