Abstract
Purpose :
Autosomal recessive Stargardt disease is the most common form of juvenile macular dystrophy and results from mutations in the ABCA4 gene. About a half of disease-causing missense changes occur in the trans-membrane or nucleotide binding domains (TM-NBD) of ABCA4. We analyzed in-silico the unfolding effects of missense changes in ABCA4 folding, stability, and association with Stargardt’s clinical phenotypes.
Methods :
The in-silico unfolding was developed to evaluate the severity of missense changes in inherited eye disease. Unfolding predictions were tested with protein crystal structures from the PDB database and experimental unfolding values from ProTherm database. The experimental results for these mutants were compared with the unfolding of protein atomic structures mutated in-silico. In-silico unfolding was applied after the validation to analyze mutant variants for ABCA4 NBD-TM domains (~1000 residues) modeled by homology and molecular dynamics. Patient genotypes came from the National Eye Institute and EyeGene (NIH), Moorfields Eye Hospital/UCL (London, UK), Catholic Medical University (Rome, Italy), and Kellogg Eye Center (Ann Arbor, MI). A group of 133 missense mutations was selected from the genotypes and mutation unfolding effects were examined. Also, the unfolding mutation scan (McCafferty & Sergeev, ARVO2016) was applied to determine crucial structural elements of NBD-TM domains which potentially susceptible to a loss of native conformation.
Results :
The average difference between experimental and in-silico unfolding values for 16 test atomic structures was ~19%, ranging from 9-35%, indicating a reasonable match of in-silico unfolding in experimental data. In the group of 133 missense mutations, 67% were associated with a loss of native conformation, 24% with the folding/unfolding equilibrium, and 9% with the protein fold stabilization. In-silico unfolding was used for the prediction of genotype severities. The in-silico severities show some association with clinical severity of retinal disease determined for 80 Stargardt’s patients from UK and 20 patients from Italy (Pearson’s r=0.68, 0.81; Adj. R-square=0.45, 0.74).
Conclusions :
Protein misfolding caused by disease-related missense change could affect structure and function of the ABCA4 protein. In-silico unfolding could be useful in establishing genotype-to-phenotype relationships in Stargardt’s disease.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.