Abstract
Purpose :
The ABCA4 protein, found in the membrane of photoreceptor outer segment discs, is essential for transport of retinoids in the visual transduction cycle. Variants (~3,000 identified) in the ABCA4 gene give rise to impaired protein function, ultimately leading to various inherited retinal dystrophies (IRDs). A key feature of ABCA4 function is its interaction with retinoids and recent studies implicate its extracellular domain (ECD2) in retinoid binding and transport. In this study, we evaluated the concordance of informatics-based pathogenicity prediction tools for 126 ABCA4 variants localized to ECD2. Further, we asked whether ABCA4 missense variants map to specific clusters within ECD2, how these residues influence retinoid interaction, and how this correlates with the 3D structure of the domain.
Methods :
ECD2 variants reported in ClinVar database were analyzed using pathogenicity prediction tools (PolyPhen2, Provean, and SIFT). Following in silico prediction, variants classified as pathogenic across all tools (PAAT) were mapped to the conserved regions or critical conserved motifs (CCMs) of the ECD2 domain. Variants in the CCMs and other implicated residues in the ABCA4 cryo-EM were expressed and functionally characterized by fluorescence-based methods. Protein modeling was used to understand the functional changes from a structural perspective.
Results :
Forty-eight percent of the 126 variants analyzed were found to be pathogenic across all tools (PAAT). Out of the total number of PAAT variants (61), 62% mapped to the CCMs. The convergence of these clusters in the cryo-EM structures also supported the significance of the CCMs in retinoid interaction, emphasizing their relevance in understanding the pathophysiology of IRDs. Fluorescence based assays demonstrated that these CCM residues (specifically those in the N-terminus of the domain) were indeed crucial for retinoid interaction.
Conclusions :
By employing in silico and function-based methods, we have presented an integrated approach to delineate the functions of ECD2. We expect that these studies will provide insight into the mechanisms of ECD2 function and, therefore, provide a larger scope on ABCA4-associated IRDs and classification of its genetic variants.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.