Abstract
Purpose:
Since retinal dystrophies (RD) are a group of blinding diseases with marked genetic and clinical heterogeneity, next generation sequencing (NGS) with its massive parallelization is currently considered the most efficient approach to identify mutations in RD genes. While targeted gene approaches neglect mutations in genes that have not been associated with a disease entity so far, whole exome sequencing (WES) has the potential to discover new disease-causing genes. We screened 24 patients with different forms of RD using WES to detect mutations in known RD genes and potentially discover novel disease genes.
Methods:
We applied WES in a cohort of 24 patients from 12 families. The patients had previously been tested negative upon screening for mutations in frequently affected genes applying Sanger sequencing, high-throughput mutation microarrays (Asper chips) and/or our NGS-based Retina-All-Panel. After stringent filtering only low-frequency (MAF = 0.01-0.05) and rare (MAF < 0.01) genetic variants remained. Prioritizing of the disease causing variant(s) was achieved on the basis of (1) prevalence in human populations, (2) predicted annotations, (3) expression of the affected gene in the retina, and (4) segregation with the disease.
Results:
In a genetically pre-selected cohort of 24 patients from 12 unrelated families, WES revealed disease causing variants in six families (11 patients) in five known RD genes (USH2A, BBS9, CRB1, CYP4V2, IFT140). In one of these families, one sibling has been diagnosed with Bardet-Biedl syndrome (BBS), while her brother has a Leber congenital amaurosis (LCA)-like phenotype. Although we were able to identify a homozygous splice site mutation in the BBS9 gene in the first sibling, the underlying genetic cause of the LCA-like phenotype in the second sibling is still unknown. Six of the 12 families remain unsolved and are subjected to further analyses (e.g. screening patient cohorts for candidate variants, whole genome sequencing to detect deep intronic variants or structural abnormalities).
Conclusions:
We performed WES in a genetically pre-selected cohort of 24 patients from 12 families with RD and were able to identify the causative mutations in 50% of cases. Compared with conventional Sanger sequencing, WES appears to be advantageous with regard to speed and cost, supporting its potential utility in molecular diagnosis in RD patients.