Abstract
Purpose :
The effect of missense mutations on protein stability is crucial in understanding inherited eye disease. Here we have created the unfolding mutation screen (UMS) for in-silico evaluation of protein stability changes and folding due to genetic perturbation.
Methods :
The atomic structures of 16 proteins used for the validation of the UMS were obtained from the RCSB database. Free energy changes (ΔΔG) for 1391 mutant variants of same proteins obtained from experimental unfolding/refolding were taken from the ProTherm database. ΔΔG for the same mutant variants were obtained in-silico using the FoldX algorithm. Both experimental and calculated ΔΔG were used to evaluate and compare the unfolding propensities of these mutations. In addition, a UMS code using Python, Bash, and R languages was created to mutate each residue of the atomic protein structure to 20 different amino acids, characterize properties of each mutation using identity and clustered unfolding heat maps, and protein structure patterned by the foldability parameter. The method was applied to several human proteins from inherited eye disease (rhodopsin, TIMP3 and REP-1), which were built by homology modeling and equilibrated in water using molecular dynamics.
Results :
In average each protein from validation dataset had 77.9±9.1% matches and an average fit score of 0.19±0.06 between the unfolding fraction of the experimental and computed values. This suggests that calculated unfolding propensities consistent with experimental unfolding values. After, ~7000 rhodopsin, ~4000 TIMP3, and ~12500 REP-1 mutations were analyzed. For rhodopsin the average unfolding fractions for class I and class II mutations were consistent with phenotype data (0.392±0.22 and 0.998±0.20 respectively). In TIMP3 the average foldability for the cysteine residues was 17.0±0.01, supporting the significance of the disulfide bonds in protein folding. Finally for REP-1 our patterned foldability structure demonstrated 4 distinct domains identified severe mutation effects due to the disruption of the hydrophobic core.
Conclusions :
UMS is a tool for predicting the unfolding effects of missense mutations on protein structure, stability, and disease phenotype. The method could be useful for express analysis of missense mutation severities and genotype-to-phenotype associations in clinical studies as well as for the analysis of the next generation sequencing data.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.