To understand how a pathogenic mutation causes inherited eye disease, it is necessary recognize how pathogenic mutations could affect protein structure-function, metabolic pathways, and how these perturbations could be associated clinical parameters describing the disease phenotype. For this purpose, we perform molecular modeling to build protein structure, evaluate the severity of genetic missense changes from the atomic level of protein, and provide a quantitative analysis of the mutation impact on protein structure, stability, and catalytic activity. We also do experimental in vitro studies for proteins of interest to measure the protein fold destabilization and changes in catalytic activity caused by the disease-related mutations. Finally, we correlate these findings with clinical phenotypes from inherited eye disease. In addition, in collaboration with the National Center for Advancing Translational Sciences (NCATS) we search for drug activators of catalytic activity of mutant protein affected by genetic mutation. This year we were using oculocutaneous albinism, Stargardts macular degeneration, Leber congenital amaurosis (LCA), and others as our disease models. We study recombinant proteins and their mutant variants in vitro to mimick patient pathogenic mutations from oculocutaneous albinism type 1 (OCA1), an autosomal recessive cause of childhood blindness. OCA1 is a rare genetic disorder of melanin synthesis that results in hypopigmented hair, skin, and eyes. Tyrosinase (TYR) catalyzes the rate-limiting, first step in melanin production and its gene is mutated in OCA1. Patients with no or reduced TYR activity are classified as OCA1A or OCA1B forms, respectively. Biochemical, biophysical, and in silico studies of human recombinant tyrosinases were performed to understand folding and stability of missense changes, which mimic genetic changes in both subtypes of OCA1 patients (Dolinska et al, PCMR, 2017). We shown that the OCA1A mutants are unstable and enzymatically inactive. In contrast, OCA1B mutants are more stable and active in varying degrees. The results are consistent with clinical data, which indicates that OCA1A mutations inactivate tyrosinase and result in severe phenotype, while OCA1B mutations partially inactivate tyrosinase and result in OCA1B albinism. These results suggest a direct link between protein stability and loss of pigmentation in OCA1B. This year we continue the search for the small drugs, which could perform as activators of human tyrosinase mutant variants. To date 34,000 compounds from the Genesis Drug Collection, the Natural Products Library, and the NCATS Pharmaceutical Collection were successfully screened using pure tyrosinase at the NCATS (NIH). The search has found new inhibitors (>100) and several activators of tyrosinase. In addition, we shown that full-length and truncated tyrosinases have similar enzymatic activities (Dolinska et al, PCMR, 2017). This study validates our drug screening, where we are using the truncated protein for a high-throughput screening. To understand experimental data and possible clinical impact of mutations we perform molecular modeling and computer simulations of protein atomic structures. We developed the unfolding mutation screen (UMS) for in silico evaluation of the severity of genetic perturbations at the atomic level of protein structure (McCafferty & Sergeev, Scientific Reports, 2016). The program considers the protein-unfolding curve and generates propensities using calculated free energy changes for every possible missense mutation. These results are presented in a series of unfolding heat maps and a colored protein 3D structure to show the residues critical to the protein folding. UMS tested with 16 crystal structures to evaluate the unfolding for 1391 mutations from the ProTherm database. Our results showed that the computational accuracy of the unfolding calculations was like the accuracy of previously published free energy changes but provided a better scale. The unfolding predictions for proteins involved in age-related macular degeneration, retinitis pigmentosa, and LCA matched well with data from previous studies. In addition, the UMS is designed to make this dataset readable for investigators with different backgrounds, who may not have any preliminary experience in homology modeling and the calculations of protein stability. We created a library of protein homology models or crystal structures and UMS maps (McCafferty & Sergeev, Scientific Data, 2016). The library is currently composed of 15 protein structures from 14 inherited eye diseases including retina degenerations, glaucoma, and cataracts, and contains data for 181,110 mutations. The UMS protein library could be used as a tool for the express analysis of novel mutations in clinical practice, next generation sequencing, and genotype-to-phenotype relationships in inherited eye disease. The COMMAD syndrome was as studied by a variety of methods including molecular modeling (George et al, AJHG, 2016). For this purpose, 26 complexes were built and refined using molecular dynamics. Two wild type complexes and eight mutant variants associated with either the MITF A isoform or the D-isoform were built as MITF: M-Box or MITF: E-Box homo-dimers. In addition,16 protein hetero-complexes were built to mimic M-box/E-box complexes formed by the MITF A-isoform: WT/R318del (father), WT/K307N (mother), K307N/R318del (son), WT/WT (daughter); or the MITF D-isoform. The Needleman-Wunsch structural alignment shows a loose structural fitting of the deletion mutant to the crystallographic template (r.m.s. of 3.2 ), compared to that of a wild type or the variant with missense mutation in both MITF-A and D isoforms (r.m.s. of 0.9-1.5 ). This suggests that deletion in position 318 causes a more significant change in the subunit structure than that of the K307N mutant. Also, initial comparison of WT and mutant MITF protein models predicts a trend of decreasing DNA binding of homodimers to the DNA in the following order: wt/wt>K307N/K307N>R318del/R318del. Moreover, the equilibration of the deletion mutant complex shows complex dissociation and complete loss of DNA binding. This suggests that deletion variants do not favor a DNA binding complex and agree with the observation that R318del variant does not bind DNA. For the WT/mutant combination, three different complexes i.e. WT+WT, WT+mutant and mutant+mutant, are possible. The binding energy, for each of these complexes was determined using molecular modelling since it could not have been done experimentally. The calculated binding energies combined for all the patient genotypes were consistent with the electrophoretic mobility shift assay data. Autosomal recessive Stargardts disease is the most common form of juvenile macular dystrophy and results from mutations in the ABCA4 gene. The atomic structure of these domains obtained using molecular modeling. The structure of transmembrane and nucleotide-binding domains of the ABCA4 protein responsible for the flipping of N-retinylidene-PE across photoreceptor disc membranes. The influence of the N1868I, G863A variants on protein structure and function were assessed by prediction from the atomic level of protein structure. In ABCA4 model, both variants, N1868I and G863A, are located on the same side the of the photoreceptor outer segment disc membrane, exposed to disk lumen, and both were predicted to be critical for the protein folding and protein stability. While not directly experimentally demonstrated, it is predicted and very plausible that both variants forming the complex allele will have a synergistic effect, which is consistent with an genetics study (Zernant et al., J. Med. Genet., 2017).
Showing the most recent 10 out of 51 publications
