Abstract
Purpose :
The ABCA4 protein, essential for vision by removing toxic retinal byproducts of phototransduction, is linked to various inherited retinal disorders (IRDs) due to its dysfunction caused by genetic variants. Functional characterization and pathogenicity assessment of the extensive number of ABCA4 variants of uncertain significance (VUS) (>900) pose a significant challenge. This study aims to functionally characterize ABCA4 missense VUS using a virus-like particle (VLP) platform and computational protein structure analysis to elucidate their impact on protein function.
Methods :
We utilized a VLP-based system for ABCA4 expression, overcoming traditional membrane protein expression limitations. We then compared the variant ABCA4-VLPs to the wild type (WT) in terms of expression level, surface localization, basal and retinal-stimulated ATPase enzymatic functions. Simultaneously, we conducted computational analyses, including protein modeling and structure analysis, to obtain multiple lines of pathogenicity evidence for ABCA4 VUS.
Results :
The VLP system facilitated large-scale expression of ABCA4, enhancing efficiency and preserving native protein structure. Biochemical assays verified the biological activity of the WT ABCA4 in VLPs. Our analyses uncovered varying degrees of changes in localization, basal and retinal-stimulated ATPase activity associated with the variants. The computational analysis identified structural features aligning with the in vitro functional assays. This integrated approach provided substantial pathogenicity evidence and aided in reclassifying some VUS according to ACMG/AMP guidelines.
Conclusions :
This study presents the reclassification of ABCA4 VUS in IRDs through combined computational and functional analyses. The in silico approaches were instrumental in prioritizing variants and pinpointing the molecular mechanisms underlying protein dysfunction, while the VLP platform offered a streamlined method for higher-throughput in vitro variant functional characterization. Understanding the effect of variation on disease and providing accurate risk assessment will be critical to implementing precision medicine. The methodology reported here can be extended beyond ABCA4 and applied to a broader range of membrane proteins associated with other genetic disorders.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.