Whole exome sequencing using Ion Proton system enables reliable genetic diagnosis of inherited retinal dystrophies

Abstract

Inherited retinal dystrophies (IRD) comprise a wide group of clinically and genetically complex diseases that progressively affect the retina. Over recent years, the development of next-generation sequencing (NGS) methods has transformed our ability to diagnose heterogeneous diseases. In this work, we have evaluated the implementation of whole exome sequencing (WES) for the molecular diagnosis of IRD. Using Ion Proton^TM system, we simultaneously analyzed 212 genes that are responsible for more than 25 syndromic and non-syndromic IRD. This approach was used to evaluate 59 unrelated families, with the pathogenic variant(s) successfully identified in 71.18% of cases. Interestingly, the mutation detection rate varied substantially depending on the IRD subtype. Overall, we found 63 different mutations (21 novel) in 29 distinct genes, and performed in vivo functional studies to determine the deleterious impact of variants identified in MERTK, CDH23, and RPGRIP1. In addition, we provide evidences that support CDHR1 as a gene responsible for autosomal recessive retinitis pigmentosa with early macular affectation, and present data regarding the disease mechanism of this gene. Altogether, these results demonstrate that targeted WES of all IRD genes is a reliable, hypothesis-free approach, and a cost- and time-effective strategy for the routine genetic diagnosis of retinal dystrophies.

Figure 1. Coverage statistics of coding regions of genes included in the panel.

The percentage of nucleotides with 0x, 1x–14x, 15x–34x, or ≥35x depth coverage per gene is shown. Black lines represent the average depth in each case.

Figure 2. Mutations identified in a cohort of 59 IRD families using a targeted WES strategy.

(a) Distribution and frequencies of IRD genes. ABCA4, USH2A and PDE6A were the most prevalent genes. (b) Types of mutations identified and their frequencies. (c) Comparison of the mutation detection rate of different IRD subtypes. Abbreviations: ACHR, achromatopsia; CD, cone dystrophy; CRD, cone–rod dystrophy; GA, gyrate atrophy; LCA, Leber congenital amaurosis; RP, retinitis pigmentosa; STGD, Stargardt disease; US, Usher syndrome.

Figure 3. Identification of CDHR1 mutations in a family affected by RP.

(a) Cosegregation analysis of CDHR1 variants identified in family Fi15/19. (b and c) Fundus eye photographs and autofluorescence images of the affected member III:2. (d) Electroretinographic recordings under scotopic (0 dB) and photopic (0 dB) conditions from both eyes of patient III:4. A b-wave could be detected in photopic ERG (amplitude and latency values are shown). Photopic 30-Hz flicker (0 dB) was also recorded. (e) RT-PCR analysis of CDHR1 blood mRNA of affected patients (III:1 and III:2), carriers (III:3, III:6, III:8, and IV:1), and one control individual (WT). Patients and carriers showed a dramatic decrease in the CDHR1 normal transcript. III:1, III:2, and III:3, who carry c.1554-2A > C, also produced a lower mass band corresponding to an mRNA that skips exon 15, whereas members III:6, III:8 and IV:1, carriers of c.1485 + 2T > C, presented a 769 bp band that directly links exon 12 and 14 of the gene.

Figure 4. In vivo functional studies of variants identified in MERTK, CDH23, and RPGRIP1 genes.

(a-i) Pedigree of consanguineous family Fi15/20, showing the cosegregation analysis of variant c.1961G > T of MERTK. (a-ii) Electrophoresis gel of the RT-PCR products obtained from affected patients (III:2 and III:4), carrier (IV:1), and WT blood samples. Patients produced high levels of an aberrantly spliced transcript, whereas the WT produced only the expected 857 bp band. (a-iii) Quantitative analysis of MERTK levels by real-time RT-PCR. The WT sample was set at 100%. (b-i) Cosegregation analysis of the two new CDH23 variants identified in family Fi15/36. (b-ii) RT-PCR assay using blood samples from the affected patient (III:2) and an unrelated WT individual. Two different bands were obtained in the patient, the expected one and another skipping exon 51. (b-iii) Chromatogram of the 2,566 bp band obtained from cDNA of patient III:2. (c-i) Fi15/12 family pedigree and RPGRIP1 variants cosegregation. (c-ii) RT-PCR analysis revealed intrafamilial differences in the expression of RPGRIP1 gene. (c-iii) Quantification of RPGRIP1 canonical transcript by real-time RT-PCR. The noncarrier family member III:3 was used as a control, and her sample was set at 100%.