Abstract
Purpose :
Stargardt disease (STGD1) is a frequent inherited retinal disease (IRD) affecting ~1/8,000 people. Significant advances have been made over the recent years in elucidating the molecular basis of STGD1, with over 600 pathogenic coding variants and with a substantial number of deep-intronic splicing variants in the disease gene ABCA4. The cis-regulatory domain of ABCA4 is unexplored so far and may represent an attractive target for non-coding disease-causing or modifying variants. By mapping and functionally validating ABCA4 putative cis-regulatory elements (CREs), we aimed to gain more insights into ABCA4 expression regulation.
Methods :
In order to identify candidate CREs, we integrated published and in-house human retinal epigenomics datasets. CRE predictions were based on chromatin accessibility (ATAC-seq), chromosome conformation capture combined with ChIP-seq (Hi-ChIP), histone modifications (ChIP-seq) and transcriptomics data (RNA-seq), all generated on human donor retina. To functionally validate in silico predicted CREs, dual luciferase reporter assays using pGL4.23 vectors were performed in hTERT RPE-1 cells. A chromatin interaction profile of the ABCA4 locus was obtained via UMI-4C experiments on neural retina and RPE from human donors.
Results :
A total of 21 predicted CREs were cloned both in their native and reverse orientation in order to assess their regulatory effect in vitro. Five regions showed an increase of reporter activity, three of which display active enhancer marks (H3K4me1 and H3K27ac) in photorreceptors, and three regions showed a significant decrease in luciferase activity. The UMI-4C data showed a decrease in background compared to previously generated 4C-seq data (Gómez-Skarmeta et al., unpblished data) and an improved sensitivity and resolution. The generation of replicates and reverse experiments to confirm the interactions are currently ongoing.
Conclusions :
Using an integrated approach based on data mining of retinal datasets, in vitro functional validation of putative retinal CREs and targeted chromosome conformation capture (UMI-4C), we have gained insight into the cis-regulatory landscape of ABCA4. The CREs identified and validated here can represent targets of non-coding pathogenic and modifying variants in cases with unsolved ABCA4-associated disease. An improved annotation of tissue-specific cis-regulatory domains of IRD genes may advance the interpretation of non-coding variants.
This is a 2021 ARVO Annual Meeting abstract.