Skip to main content
NCBI home page
JAMA Network logoLink to JAMA Network
. 2021 Jan 28;139(3):319–328. doi: 10.1001/jamaophthalmol.2020.6418

Analysis of Pathogenic Variants Correctable With CRISPR Base Editing Among Patients With Recessive Inherited Retinal Degeneration

Lewis E Fry 1,2, Michelle E McClements 1,2, Robert E MacLaren 1,2,✉
PMCID: PMC7844696  PMID: 33507217

Key Points

Question

What is the prevalence of pathogenic single-nucleotide variants associated with inherited retinal disease that could be repaired through gene editing with base editors?

Findings

In this cross-sectional study of 12 369 alleles and 179 patients included in the genetic databases of large genes (ABCA4, CDH23, MYO7A, CEP290, USH2A, and EYS) implicated in retinal degeneration, 53% of pathogenic alleles were editable, and 76% of patients who received diagnoses through a genetic service had an editable allele.

Meaning

The study’s results indicate that a substantial proportion of patients with inherited retinal disease could be treated with base editing and suggest that therapeutic strategies focused on common variants could be used to treat a large number of patients.

Abstract

Importance

Many common inherited retinal diseases are not easily treated with gene therapy. Gene editing with base editors may allow the targeted repair of single-nucleotide transition variants in DNA and RNA. It is unknown how many patients have pathogenic variants that are correctable with a base editing strategy.

Objective

To assess the prevalence and spectrum of pathogenic single-nucleotide variants amenable to base editing in common large recessively inherited genes that are associated with inherited retinal degeneration.

Design, Setting, and Participants

In this retrospective cross-sectional study, nonidentifiable records of patients with biallelic pathogenic variants of genes associated with inherited retinal degeneration between July 2013 and December 2019 were analyzed using data from the Oxford University Hospitals Medical Genetics Laboratories, the Leiden Open Variation Database, and previously published studies. Six candidate genes (ABCA4, CDH23, CEP290, EYS, MYO7A, and USH2A), which were determined to be the most common recessive genes with coding sequences not deliverable in a single adeno-associated viral vector, were examined. Data were analyzed from April 16 to May 11, 2020.

Main Outcomes and Measures

Proportion of alleles with a pathogenic transition variant that is potentially correctable with a base editing strategy and proportion of patients with a base-editable allele.

Results

A total of 12 369 alleles from the Leiden Open Variation Database and 179 patients who received diagnoses through the genetic service of the Oxford University Hospitals Medical Genetics Laboratories were analyzed. Editable variants accounted for 53% of all pathogenic variants in the candidate genes contained in the Leiden Open Variation Database. The proportion of pathogenic alleles that were editable varied by gene; 63.1% of alleles in ABCA4, 62.7% of alleles in CDH23, 53.8% of alleles in MYO7A, 41.6% of alleles in CEP290, 37.3% of alleles in USH2A, and 22.2% of alleles in EYS were editable. The 5 most common editable pathogenic variants of each gene accounted for a mean (SD) of 19.1% (9.5%) of all pathogenic alleles within each gene. In the Oxford cohort, 136 of 179 patients (76.0%) had at least 1 editable allele. A total of 53 of 107 patients (49.5%) with biallelic pathogenic variants in the gene ABCA4 and 16 of 56 patients (28.6%) with biallelic pathogenic variants in the gene USH2A had 1 of the 5 most common editable alleles.

Conclusions and Relevance

This study found that pathogenic variants amenable to base editing commonly occur in inherited retinal degeneration. These findings, if generalized to other cohorts, provide an approach for developing base editing therapies to treat retinal degeneration not amenable to gene therapy.

This cross-sectional study uses data from the Oxford Medical Genetics Laboratory, the Leiden Open Variation Database, and previous studies to assess pathogenic single-nucleotide variants amenable to base editing in common recessive genes associated with inherited retinal degeneration.

Introduction

Monogenic inherited retinal degeneration (IRD) is associated with pathogenic variants in genes associated with outer retinal cells and patients with the disease experience substantial visual loss. Gene therapy to treat IRD by providing a normal copy of the relevant gene is now an approved treatment for IRD associated with the retinoid isomerohydrolase gene, RPE65 (OMIM 180069),1 and is being examined in clinical trials for many other genes.2 Adeno-associated viral (AAV) vectors are popular for retinal gene therapy, as they can efficiently target retinal cells with low immunogenicity and have an established clinical safety profile.3 However, the genes implicated in many common recessively inherited IRDs are not easily treatable with AAV gene therapy, as the coding sequences often exceed the 4.7 kilobase capacity of a single AAV vector.

A potential therapeutic alternative for these large genes is the use of gene editing to correct the endogenous genetic sequence. Tools such as the clustered regularly interspaced short palindrome repeats (CRISPR)–associated protein 9 (Cas9) system enable the precise and efficient alteration of nucleic acids.4 Base editors are a class of gene editing tool that directly create point variants at specified loci in DNA or RNA.5

DNA base editors are capable of repairing all 4 transition variants (C>T, G>A, T>C, and A>G) in DNA.5 The current generation of CRISPR-Cas base editors predominantly use a Cas9 nuclease fused to a deaminase domain to mediate base conversion through hydrolytic deamination. The base editor is directed to the specific loci to be edited by a guide RNA complementary to the target region.5 Cytosine base editors convert a C-G into a T-A base pair through a fused cytidine deaminase and a uracil DNA glycosylase inhibitor. Adenine base editors convert an A-T to a G-C base pair using a fused engineered Escherichia coli transfer RNA adenosine deaminase enzyme, TadA. Unlike other gene correction methods, base editing does not require the creation of double-stranded breaks in DNA6,7 or require a donor template of DNA for repair.8,9

RNA base editors can correct 2 genomic transition variants in RNA (G>A and T>C).10 These base editors use an RNA-specific adenosine deaminase (ADAR) to convert adenosine to inosine, which is read as guanosine by cellular machinery, effectively mediating an A-G change in messenger RNA.10 The ADAR protein or its deaminase domain can be recruited to the site of interest by using antisense methods11,12 or by fusing the deaminase domain to an RNA-binding protein, such as deactivated Cas13.12,13,14 Engineered ADARs can also mediate cytosine to uracil conversions.15 As RNA is single stranded, together these techniques allow the directed correction of G>A and T>C genomic variants in RNA. Because RNA editing occurs at a transcript level, this approach reduces the risk of creating permanent off-target changes in the genome.10

In this study, we sought to identify the primary genes for which base editing would benefit, and the spectrum of pathogenic variants within these that could be targeted. Based on 4 large cohorts of 6986 total patients with IRD,16,17,18,19 we identified 6 of the most commonly occurring recessive genes with long coding sequences in patients with a molecular diagnosis (Table 1). These genes included ATP-binding cassette, subfamily A, member 4 (ABCA4; OMIM 601691; NM_000350.2; mean [SD] prevalence, 20.0% [2.4%]); Usher syndrome, type 2A (USH2A; OMIM 276901; NM_206933.2; mean [SD] prevalence, 8.0% [2.0%]); myosin 7A (MYO7A; OMIM 276903; NM_000260.3; mean [SD] prevalence, 1.6% [0.4%]); eyes shut homologue (EYS; OMIM 612424; NM_001142800.1; mean [SD] prevalence, 2.5% [1.8%]); centrosomal protein, 290-KD (CEP290; OMIM 610142; NM_025114.3; mean [SD] prevalence, 2.3% [2.3%]); and Usher syndrome, type 1D (CDH23; OMIM 601067; NM_022124.5; mean [SD] prevalence, 0.5% [0.1%]). These genes cumulatively accounted for 35% of all individuals with an IRD in these studies.

Table 1. Prevalence of Large Recessive Genes in Large Published Cohorts of Individuals With Molecular Diagnosis of Inherited Retinal Diseasea.

Gene Range of phenotypes CDS, kilobase Prevalence, No. (%) Prevalence, mean (SD), %
UK16 (n = 4236) US17 (n = 760) Brazil18 (n = 400) Israel19 (n = 1590)b
ABCA4 Stargardt macular dystrophy; cone-rod dystrophy 6.8 789 (18.6) 173 (22.8) 84 (21.0) 278 (17.5) 20.0 (2.4)
USH2A RP; Usher syndrome, type 2 15.6 342 (8.1) 76 (10.0) 21 (5.3) 133 (8.4) 8.0 (2.0)
MYO7A Usher syndrome, type 1 6.6 58 (1.4) 8 (1.1) 8 (2.0) 29 (1.8) 1.6 (0.4)
EYS RP 9.4 43 (1.0) 6 (0.8) 16 (4.0) 64 (4.0) 2.5 (1.8)
CEP290 LCA; Joubert syndrome; ciliopathies 7.4 35 (0.8) 18 (2.4) 22 (5.5) 6 (0.4) 2.3 (2.3)
CDH23 Usher syndrome, type 1 10.1 20 (0.5) 4 (0.5) 0 1 (0.1) 0.5 (0.1)
All genes NA NA 1287 (30.4) 285 (37.5) 151 (37.8) 511 (32.1) 34.6 (3.7)
Open in a new tab

Abbreviations: CDS, coding sequence length; LCA, Leber congenital amaurosis; NA; not applicable; RP, retinitis pigmentosa; UK, United Kingdom.

a

Table includes only the genes examined in this study.

b

n = 1590 indicates families rather than individuals.

It is currently unknown how many pathogenic variants in these genes might be editable (ie, how many are transition variants that could potentially be corrected with a base editor) and what the most commonly occurring editable alleles are for each gene. It is also unknown how many patients might have editable alleles. Using data from the Leiden Open Variation Database (LOVD) and from patients with IRD who received diagnoses through the genetic services of tertiary referral hospitals in the United Kingdom and the US, we investigated the proportion of patients with a recessive IRD gene that had a pathogenic variant correctable with base editing and explored the prevalence of common target variants.

Methods

Data from patients with biallelic pathogenic variants of genes associated with inherited retinal degeneration between July 2013 and December 2019 were obtained from the Oxford University Hospitals Medical Genetics Laboratories, the LOVD, and previously published studies. The analysis was conducted using a list of previously collected nonidentifiable variants that were identified retrospectively and LOVD data that were publicly available. Therefore, research ethics committee review was not required per Oxford University guidelines for clinical research studies limited to the use of previously collected nonidentifiable information. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cross-sectional studies.

LOVD Search and Variant Coding

Variants included in the inherited retinal disease gene variant data set of the LOVD, version 3.0,20,21 were downloaded on September 6, 2019, for the following genes: ABCA4, CEP290, CDH23, EYS, MYO7A, and USH2A. The LOVD data set for the CEP290 gene contains records derived from CEP290base, a locus-specific variant data set that links variants with patients and their phenotypes.22,23 Data from CEP290base were filtered for pathogenic variants only and downloaded. Records in the CEP290base and LOVD CEP290 databases were combined, and duplicate entries were excluded to yield a single data set.

The LOVD records were cleaned and coded for pathogenicity and variational consequence (eFigure in the Supplement). For instances in which variant pathogenicity was not annotated, variants were cross-referenced with the pathogenicity comprehensively labeled in Messchaert et al24 (for the EYS gene), Cornelis et al25 (for the ABCA4 gene), and CEP290base (for the CEP290 gene). The USH2A, MYO7A, and CDH23 gene databases were complete in annotations for pathogenicity. For instances in which uncertain variant pathogenicity remained, we reviewed the variant using the ClinVar database26,27 and literature sources, and we excluded the variant from analysis if its pathogenicity could not be determined. Variants that were benign, likely benign, and of unknown significance were excluded to yield a final data set of pathogenic or likely pathogenic variants.

Each variant was categorized by expected variant consequence into termination, missense, synonymous, untranslated region, intronic, splice, insertion/duplication/deletion, or unknown. Termination was defined as a single-nucleotide variant (SNV) generating a stop codon, splice was defined as a variant disrupting a canonical splice donor (+1 and +2) or acceptor (−2 and −1) site, and intronic was defined as a variant occurring between the intronic +3 to −3 positions. Complex alleles with more than 1 pathogenic variant were classified as uneditable.

Analysis of Oxford Data

Next-generation sequencing panel testing was performed as indicated by the clinical phenotype between July 2013 and November 2019 by the Oxford University Hospitals Medical Genetics Laboratories among patients referred for suspected IRD, as previously described elsewhere.28 Assignment of pathogenicity was based on American College of Medical Genetics guidelines.29 Variants were classified as highly likely to be pathogenic (class 5), likely to be pathogenic (class 4), or of uncertain significance (class 3). Records were filtered to yield only patients with biallelic class 4 or class 5 variants in genes ABCA4, CEP290, CDH23, EYS, MYO7A, and USH2A.

The variation consequence of each variant was labeled as described for the LOVD analysis. Complex alleles with more than 1 variant were classified as uneditable. These methods were also used for analysis of the Stone et al17 data.

Statistical Analysis

Data were analyzed using Prism, version 8 (GraphPad), and Excel (Microsoft Corp) software. Data were reported as descriptive statistics, and means were reported with SDs, as appropriate. Data analysis was performed from April 16 to May 11, 2020.

Results

Prevalence of Editable Variants in LOVD

To assess the prevalence of alleles with editable pathogenic variants in the 6 most common large recessive genes, we analyzed data from the LOVD, which contains curated repositories of sequenced variants in the alleles of patients with IRD. For the genes examined (ABCA4, CEP290, CDH23, EYS, MYO7A, and USH2A), the LOVD contained a total of 26 418 alleles. Because base editing strategies focus on pathogenic variants, we excluded benign variants and variants of uncertain significance, which left a data set of 12 369 pathogenic or likely pathogenic alleles for further analysis.

Editable transition variants represented 53.0% of all pathogenic variants in all genes, with the remainder comprising noneditable variants, such as transversion SNVs, complex alleles, insertions, duplications, or deletions (Figure 1; eTable 1 in the Supplement). Variants G>A (23.1%) and C>T (18.1%) were the most prevalent types, with fewer T>C (7.0%) and A>G (4.8%) variants.

Figure 1. Distribution of Alleles From the Leiden Open Variation Database.

Figure 1.

Open in a new tab

Analysis of 12 369 individual pathogenic or likely pathogenic alleles in the Leiden Open Variation Database across the 6 most common large recessive genes associated with inherited retinal degeneration (Table 1). Alleles with transition variants targetable with base editing (G>A, C>T, T>C, A>G) are common across all genes. Other variants are not targetable with base editing, including transversion variants, insertions, duplications, deletions, and complex alleles. Additional details are available in eTable 1 in the Supplement.

Within each gene, the proportion of editable variants varied. Most of the alleles in genes ABCA4 (63.1%), CDH23 (62.7%), and MYO7A (53.8%) were editable transition variants, while a lower proportion of variants were editable in genes CEP290 (41.6%), USH2A (37.3%), and EYS (22.2%) (Figure 2A).

Figure 2. Common Editable Alleles and Variation Types From the Leiden Open Variation Database.

Figure 2.

Open in a new tab

A, Proportion of all editable pathogenic alleles. Overall proportions of each editable gene and proportion of pathogenic alleles formed from the 5 most common alleles occurring in each gene. The 5 most common editable alleles for each gene (eTable 2 in the Supplement) comprised a mean (SD) of 19.1% (9.5%) of all pathogenic variations. B, Variations by consequence. Variations of pathogenic alleles analyzed for genes ABCA4 (6540 alleles), CDH23 (609 alleles), CEP290 (483 alleles), EYS (589 alleles), MYO7A (1005 alleles), and USH2A (3142 alleles). Termination variants were defined as single-nucleotide variants that are associated with an in-frame codon being altered to a stop codon. Splicing variants were defined as those affecting the canonical positions (+1, +2, −1, and −2) of the donor or acceptor splice sites. Intronic variants were defined as those occurring in intronic regions beyond the canonical sites.

We identified the 5 pathogenic variants in each gene that occurred most frequently (Table 2; eTable 2 in the Supplement). The 5 most common editable variants accounted for a mean (SD) of 19.1% (9.5%) of all pathogenic alleles within each gene; in genes ABCA4 (21.1%), CEP290 (29.4%), CDH23 (28.9%), and MYO7A (19.1%), the 5 most common editable variants accounted for a substantial proportion of all alleles (Figure 2A). Missense variants were the predominant editable variants observed in gene ABCA4 (45.5%), and these variants were also common in genes CDH23 and MYO7A (Figure 2B).

Table 2. Comparison of 5 Most Frequent Variants in Genes ABCA4 and USH2Aa.

Gene Data set
LOVD Oxford Stone et al17
cDNA change Alleles, % cDNA change Alleles, % cDNA change Alleles, %
ABCA4 c.5882G>A 9.6 c.5882G>A 9.3 c.5461-10T>C 8.0
c.5461-10T>C 4.1 c.5461-10T>C 6.5 c.5882G>A 6.7
c.6079C>T 2.5 c.6079C>T 6.5 c.4139C>T 4.0
c.4469G>A 2.4 c.4139C>T 5.1 c.3113C>T 4.0
c.5714 + 5G>A 2.4 c.5714 + 5G>A 4.7 c.5714 + 5G>A 3.7
Subtotal NA 21.1 NA 32.2 NA 26.3
USH2A c.11864G>A 3.4 c.10073G>A 6.3 c.11864G>A 5.0
c.7595-2144A>G 1.2 c.10342G>A 3.6 c.1606T>C 2.1
c.1876C>T 1.1 c.12574C>T 2.7 c.10073G>A 2.1
c.10712C>T 1.1 c.3407G>A 2.7 c.7595-2144A>G 2.1
c.9799T>C 1.0 c.8740C>T 1.8 c.12575G>A 1.4
Subtotal NA 7.8 NA 17.0 NA 12.9
Open in a new tab

Abbreviations: cDNA, complementary DNA; LOVD, Leiden Open Variation Database; NA, not applicable; Oxford, Oxford University Hospitals Medical Genetics Laboratories.

a

The 5 most frequent alleles in each data set for genes ABCA4 and USH2A are listed in descending frequency for each data set. The proportion of alleles is expressed as a percentage of the total alleles for genes ABCA4 and USH2A in the LOVD (6540 alleles and 3143 alleles, respectively), Oxford (214 alleles and 112 alleles, respectively), and Stone et al17 (300 alleles and 140 alleles, respectively) data sets. Additional data are available in eTable 2, eTable 4, and eTable 5 in the Supplement.

Variants associated with nonsense mediated decay of the transcript, such as premature termination variants and variants that disrupt splicing, are expected to be particularly amenable to treatment, as correction is likely to rescue a null allele. Editable termination variants (defined as SNVs that are associated with in-frame codons being altered to stop codons) accounted for a mean (SD) of 12.6% (4.7%) of variants across all genes (Figure 2B). Common termination variants were found in genes USH2A, MYO7A, and EYS (eTable 2 in the Supplement). Editable variants in the canonical donor or acceptor splice sites (defined as splice variants in this study) accounted for a mean (SD) of 5.9% (4.2%) in each gene and were comparatively more prevalent in genes CDH23, USH2A, and MYO7A. Intronic variants, such as deep intronic splice variants and noncanonical splice site variants, accounted for a mean (SD) of 8.3% (9.6%) in each gene and were notably predominant in gene CEP290 (24%). Deep intronic variants were most common in both CEP290 and CHD23 genes.

Patients With Editable Variants

Many patients with IRD are heterozygous, and correction of only one of these alleles should theoretically be sufficient to ameliorate the condition in many cases. To ascertain the number of potentially treatable patients within a typical large ophthalmology genetic service data set, we analyzed the records of patients who received genetic screening for a suspected IRD at the Oxford University Hospitals Medical Genetics Laboratories. These records were filtered to include only patients with biallelic pathogenic variants in the 6 most common recessive genes, yielding a data set of 179 patients with 358 alleles. The distribution of alleles within the Oxford data set was comparable to that of the LOVD data set (Figure 3A; eTable 3 in the Supplement). Similar to alleles in the LOVD data set, 204 of 358 alleles (57.0%) were editable, with G>A (83 of 358 alleles [23.2%]) and C>T (84 of 358 alleles [23.5%]) variants the most common.

Figure 3. Analysis of Patients With Inherited Retinal Degeneration Associated With Large Recessive Genes.

Figure 3.

Open in a new tab

A, Proportion of editable variants from Oxford data set. Editable variants in 358 alleles of 179 patients with 2 confirmed pathogenic or likely pathogenic variants (American College of Medical Genetics class 4 or 5) in the 6 most common recessive genes. B, Proportion of editable variants from Stone et al17 data set. Editable variants in 255 patients with 510 alleles in the 6 most common recessive genes. C, Patients in Oxford data set by implicated gene. For genes ABCA4 and USH2A, patients were classified as having a common editable variant if they had 1 of the 5 most common variants (detailed list of variants available in Table 2 and eTable 3 in the Supplement). Among 179 total patients analyzed, 136 patients (76.0%) had an editable variant, and 42 patients (24.5%) did not have an editable variant. D, Patients in Stone et al17 data set by implicated gene. For genes ABCA4 and USH2A, patients were classified as having a common editable variant if they had 1 of the 5 most common variants (detailed list of variants available in Table 2 and eTable 3 in the Supplement). Among 255 total patients analyzed, 205 patients (80.4%) had an editable variant, and 50 patients (19.6%) did not have an editable variant. Oxford indicates Oxford University Hospitals Medical Genetics Laboratories.

We classified patients as having editable variants if they had at least 1 editable allele. Among 179 total patients in the Oxford data set, 136 individuals (76.0%) had editable variants, and 42 individuals (23.5%) did not have editable variants (Figure 3B). Individuals with pathogenic variants in genes ABCA4 (107 of 179 individuals [59.8%]) and USH2A (56 of 179 individuals [31.3%]) represented the largest proportion of patients in the Oxford data set; therefore, further analysis of the most prevalent variants in genes ABCA4 and USH2A was performed. Among 107 patients with pathogenic variants in gene ABCA4, 95 individuals (88.8%) had an allele with an editable transition variant; among 56 patients with pathogenic variants in gene USH2A, 32 individuals (57.1%) had an editable allele (Figure 3C). We then looked at the 5 most frequent editable alleles in the Oxford ABCA4 and USH2A data sets and designated common variants for these genes (Table 2; eTable 3 in the Supplement). The 5 most frequent editable alleles collectively accounted for 32.2% of ABCA4 alleles and 17.0% of USH2A alleles. Because most patients are heterozygous, 53 of 107 patients (49.5%) with pathogenic variants in gene ABCA4 and 16 of 56 patients (28.6%) with pathogenic variants in gene USH2A had at least one of these common variants (Figure 3C), suggesting that 5 guide RNAs (one for each pathogenic variant) could be used to treat a substantial proportion of patients.

To validate this finding, we performed an identical analysis of data from a similar genetic service in the US, which was previously published by Stone et al17 (Figure 3D). Among 255 patients with 510 alleles in the 6 most common recessive genes, 205 individuals (80.4%) had editable variants, and 50 individuals (19.6%) did not have editable variants. Among those with pathogenic variants in genes ABCA4, 139 of 150 patients (92.7%) had an allele with an editable transition variant, while 44 of 70 patients (62.9%) with pathogenic variants in gene USH2A had an allele with an editable transition variant (Figure 3). In the Stone et al17 data set, the 5 most frequent common variants in genes ABCA4 and USHA accounted for 26.3% and 12.9% of alleles and were present in 71 of 150 patients (47.3%) and 16 of 70 patients (22.9%) with these genes, respectively (Figure 3 and Table 2; eTable 5 in the Supplement). These data from a geographically separate population are consistent with the findings from our UK-based data set.

Across the 3 data sets analyzed, the most frequent of the 5 variants in genes ABCA4 and USH2A were compared (Table 2). In gene ABCA4, 3 variants (c.5882G>A, c.5461-10T>C, and c.5714 + 5G>A) were present in the 5 most frequent variants of each data set. In gene USH2A, no single variant was present in the 5 most frequent variants of all data sets. Three variants (c.11864G>A, c.7595-2144A>G, and c.10073G>A) occurred in 2 of the 3 data sets.

Discussion

Recent developments make the correction of pathogenic variants in DNA or RNA using base editors an exciting therapeutic prospect for genetic diseases, including IRD. Base editors have been observed to possess in vivo activity in photoreceptors30 and retinal pigment epithelium.31 Efforts to enable the delivery of DNA base editors using single or dual AAV vector approaches,30,32,33 lentiviral vectors,31 or nonviral delivery34,35,36,37 continue. Many RNA editors are deliverable within a single AAV12,38,39,40 and have indicated promise in vivo.12 The targeting range of base editors has expanded rapidly; therefore, in this study, we considered all pathogenic transition variants as potential targets. For RNA base editing with Cas13 or antisense methods, there are few known targeting restrictions, although some modalities have local nucleotide preferences.10,11,12,13 For DNA base editing, Cas enzymes with flexible or alternative protospacer-adjacent motif sequences have enabled targeting of almost any sequence in the genome.41,42,43 Molecular engineering has enhanced the on-target efficiency of base editors44,45,46 and continues to improve the editing specificity.13,44,47,48,49

We identified the spectrum of potential base editing targets across the 6 most common recessive genes that are implicated in IRD and are too large for delivery within a single AAV. Recessive diseases are ideal targets, as it is likely that a base editing strategy to correct only 1 of 2 alleles is required in many cases. More than one-half of the pathogenic variants in genes ABCA4, CEP290, CDH23, EYS, MYO7A, and USH2A are associated with transition SNVs that are editable with the current classes of base editors. This finding is consistent across the global LOVD data set and patient data from 2 large ophthalmic genetic services in the United Kingdom and the US. Although transitions account for only 4 of the possible 12 SNVs, they are more prevalent than transversions throughout the human genome.50 The GC-AT base pair transition is particularly common because of spontaneous deamination secondary to oxidative DNA damage51,52 and, consistent with this finding, 41.3% of pathogenic variants in the LOVD data set were G>A and C>T transitions in the studied genes. Because of the frequency of transition variants, we found that, in 2 separate cohorts, 76% to 80% of patients with disease associated with biallelic pathogenic variants in the 6 genes analyzed had at least 1 editable allele.

A limitation of base editing strategies is that they are variant dependent; that is, a guide RNA is required for each individual variant. Although many heterogenous private variants occur in each gene, analysis of the LOVD data indicates that targeting the 5 most common editable variants allows for approximately one-fifth of pathogenic variants in each gene to be targeted.

Pathogenic variants in gene ABCA4 are commonly associated with Stargardt disease (OMIM 248200). The most common variant is c.5882G>A; p.(Gly1961Glu), which emerged in Eastern Africa and spread throughout the global population.53 The c.5461-10T>C; p.([Thr1821Valfs, Thr1821Aspfs]) variant is a common noncanonical splice variant that is present predominantly in European populations.54 Although most variants in the ABCA4 gene are SNVs, a substantial proportion (approximately 10%)53 are complex alleles, which contain 2 variants.25 The presence of more than 1 variant would be challenging to correct with a base editing approach; thus, complex alleles were not considered editable variants in this study.

Pathogenic variants in the USH2A gene can present as nonsyndromic retinitis pigmentosa or as Usher syndrome, type 2 (USH2; OMIM 608400), which is associated with retinitis pigmentosa and hearing loss. A smaller proportion of pathogenic variants in gene USH2A were identified as editable (7.8%), and the 2 most common variants (c.2299delG; p.[Glu767fs] and c.2276G>T; p.[Cys759Phe] in exon 13) accounted for approximately 35% of cases.55,56,57 Although neither variant is amenable to base editing, inducing exon-skipping using antisense oligonucleotides,58,59 using base editing to alter splice acceptor sites,60 or using CRISPR-Cas13 mediated splice site repression61 are alternative approaches. For direct base editing, the c.11864G>A; p.(Trp3955*) variant was the most common target and appears to be prevalent in European populations.62 Data from translation read-through strategies suggest that even partial correction of nonsense targets may be sufficient to rescue the phenotype.63

Although genes CEP290, MYO7A, and CDH23 are less frequently implicated in IRD, we identified prevalent editable variants in these genes. The common deep intronic variant in gene CEP290, c.2991 + 1655A>G; p.(Cys998*),64 is an editable variant associated with Leber congenital amaurosis that creates a new intronic splice donor site. Approaches to correct this variant are currently being assessed in clinical trials.7,65,66

Loss-of-function variants in the MYO7A gene are implicated in Usher syndrome, type 1 (OMIM 276900),67 and more than 50% of MYO7A alleles in the LOVD data set were editable, with almost 15% of alleles accounted for by 3 pathogenic variants. Variants of the CDH23 gene are also associated with Usher syndrome, type 1,68,69,70 and a high proportion (62.7%) of the alleles were editable, with 38.7% of editable alleles associated with variants that were expected to be protein truncating (splice, intronic, or termination). The noncanonical splice site variant, c.6050-9G>A, was a prevalent editable variant (13.5%). This variant induces aberrant splicing with a frame shift, producing nonsense decay of the transcript.71

Pathogenic variants in gene EYS underlie 5% to 10% of nonsyndromic retinitis pigmentosa; however, few variants are editable in the alleles included in the LOVD to date. The data set contains an array of private variants, with a high proportion of insertions, duplications, and deletions, comprising 51.6% of all variants.

Translation of genome editing strategies warrants consideration of important safety issues, including off-target editing, protein immunogenicity, and delivery concerns.72 Base editing tools complement a rapidly expanding group of molecular therapies. Alternative strategies include the use of antisense oligonucleotides for the suppression of pathogenic deep intronic splice sites or the skipping of exons that contain common pathogenic variants.73 Data from antisense oligonucleotide therapy for the CEP290 gene indicated that small RNAs delivered intravitreally are able to target photoreceptors and improve the disease phenotype in humans.66 Antisense oligonucleotides were recently reported to recruit endogenously expressed ADAR proteins to target pathogenic variants with base editing.11,12 This approach may be a promising method for combining antisense and base editing strategies. For variants not amenable to base editing, other CRISPR-based strategies, such as prime editing and homology-directed repair, may be applicable.72

Limitations

This study has several limitations. Because the data sets were not reliably coded with ethnicity or demographic information, our analysis does not take into account sampling biases and differences in the population prevalence of certain variants or genes. In addition, because we sought to identify targets that are known to be associated with disease, the study only considers definitively pathogenic variants and not variants of unknown significance.

Conclusions

We present a rationale and approach for potential base editing targets in the retina from 3 data sets. The study’s findings support the conclusion that base editing is a promising tool with many potential therapeutic targets, in an increasing array of gene editing techniques that might be developed for patients with inherited retinal disease due to variants in large genes that are not readily treatable with AAV gene therapy.

Supplement.

eTable 1. Pathogenic Alleles From the LOVD Broken Down by Variation Type and Gene, as Analyzed in Figure 1

eTable 2. Most Common Variants in the LOVD

eTable 3. Pathogenic Alleles From the Oxford Database Broken Down by Variation Type and Gene, as Analyzed in Figure 3A

eTable 4. Summary of the 5 Most Common Alleles for Genes ABCA4 and USH2A in the Oxford Data Set

eTable 5. Summary of the 5 Most Common Alleles for Genes ABCA4 and USH2A in the Stone et al17 Data Set

eFigure. Flow Diagram Outlining Method of LOVD Analysis

Click here for additional data file. (442.6KB, pdf)

References

  • 1.Russell S, Bennett J, Wellman JA, et al. Efficacy and safety of voretigene neparvovec (AAV2-hRPE65v2) in patients with RPE65-mediated inherited retinal dystrophy: a randomised, controlled, open-label, phase 3 trial. Lancet. 2017;390(10097):849-860. doi: 10.1016/S0140-6736(17)31868-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Vazquez-Dominguez I, Garanto A, Collin RWJ. Molecular therapies for inherited retinal diseases—current standing, opportunities and challenges. Genes (Basel). 2019;10(9):654. doi: 10.3390/genes10090654 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Lee JH, Wang J-H, Chen J, et al. Gene therapy for visual loss: opportunities and concerns. Prog Retin Eye Res. 2019;68:31-53. doi: 10.1016/j.preteyeres.2018.08.003 [DOI] [PubMed] [Google Scholar]
  • 4.Cong L, Ran FA, Cox D, et al. Multiplex genome engineering using CRISPR/Cas systems. Science. 2013;339(6121):819-823. doi: 10.1126/science.1231143 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Rees HA, Liu DR. Base editing: precision chemistry on the genome and transcriptome of living cells. Nat Rev Genet. 2018;19(12):770-788. doi: 10.1038/s41576-018-0059-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Tsai Y-T, Wu W-H, Lee T-T, et al. Clustered regularly interspaced short palindromic repeats-based genome surgery for the treatment of autosomal dominant retinitis pigmentosa. Ophthalmology. 2018;125(9):1421-1430. doi: 10.1016/j.ophtha.2018.04.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Maeder ML, Stefanidakis M, Wilson CJ, et al. Development of a gene-editing approach to restore vision loss in Leber congenital amaurosis type 10. Nat Med. 2019;25(2):229-233. doi: 10.1038/s41591-018-0327-9 [DOI] [PubMed] [Google Scholar]
  • 8.Suzuki K, Tsunekawa Y, Hernandez-Benitez R, et al. In vivo genome editing via CRISPR/Cas9 mediated homology-independent targeted integration. Nature. 2016;540(7631):144-149. doi: 10.1038/nature20565 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Cai Y, Cheng T, Yao Y, et al. In vivo genome editing rescues photoreceptor degeneration via a Cas9/RecA-mediated homology-directed repair pathway. Sci Adv. 2019;5(4):eaav3335. doi: 10.1126/sciadv.aav3335 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Fry LE, Peddle CF, Barnard AR, McClements ME, MacLaren RE. RNA editing as a therapeutic approach for retinal gene therapy requiring long coding sequences. Int J Mol Sci. 2020;21(3):777. doi: 10.3390/ijms21030777 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Qu L, Yi Z, Zhu S, et al. Programmable RNA editing by recruiting endogenous ADAR using engineered RNAs. Nat Biotechnol. 2019;37(9):1059-1069. doi: 10.1038/s41587-019-0178-z [DOI] [PubMed] [Google Scholar]
  • 12.Katrekar D, Chen G, Meluzzi D, et al. In vivo RNA editing of point mutations via RNA-guided adenosine deaminases. Nat Methods. 2019;16(3):239-242. doi: 10.1038/s41592-019-0323-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Cox DBT, Gootenberg JS, Abudayyeh OO, et al. RNA editing with CRISPR-Cas13. Science. 2017;358(6366):1019-1027. doi: 10.1126/science.aaq0180 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Sinnamon JR, Kim SY, Corson GM, et al. Site-directed RNA repair of endogenous Mecp2 RNA in neurons. Proc Natl Acad Sci U S A. 2017;114(44):E9395-E9402. doi: 10.1073/pnas.1715320114 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Abudayyeh OO, Gootenberg JS, Franklin B, et al. A cytosine deaminase for programmable single-base RNA editing. Science. 2019;365(6451):382-386. doi: 10.1126/science.aax7063 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Pontikos N, Arno G, Jurkute N, et al. Genetic basis of inherited retinal disease in a molecularly characterised cohort of over 3000 families from the United Kingdom. Ophthalmology. 2020;127(10):1384-1394. doi: 10.1016/j.ophtha.2020.04.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Stone EM, Andorf JL, Whitmore SS, et al. Clinically focused molecular investigation of 1000 consecutive families with inherited retinal disease. Ophthalmology. 2017;124(9):1314-1331. doi: 10.1016/j.ophtha.2017.04.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Motta FL, Martin RP, Filippelli-Silva R, Salles MV, Sallum JMF. Relative frequency of inherited retinal dystrophies in Brazil. Sci Rep. 2018;8(1):15939. doi: 10.1038/s41598-018-34380-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Sharon D, Ben-Yosef T, Goldenberg-Cohen N, et al. A nationwide genetic analysis of inherited retinal diseases in Israel as assessed by the Israeli inherited retinal disease consortium (IIRDC). Hum Mutat. 2020;41(1):140-149. doi: 10.1002/humu.23903 [DOI] [PubMed] [Google Scholar]
  • 20.Fokkema IFAC, Taschner PEM, Schaafsma GCP, Celli J, Laros JFJ, den Dunnen JT. LOVD v.2.0: the next generation in gene variant databases. Hum Mutat. 2011;32(5):557-563. doi: 10.1002/humu.21438 [DOI] [PubMed] [Google Scholar]
  • 21.Leiden Open Variation Database . Version 3.0. Leiden University Medical Center. Accessed September 6, 2019. https://databases.lovd.nl/shared/genes/
  • 22.Coppieters F, Lefever S, Leroy BP, De Baere E. CEP290, a gene with many faces: mutation overview and presentation of CEP290base. Hum Mutat. 2010;31(10):1097-1108. doi: 10.1002/humu.21337 [DOI] [PubMed] [Google Scholar]
  • 23.CEP290base. Locus-specific variant database. Updated October 2009. Accessed April 10, 2020. https://cep290base.cmgg.be/
  • 24.Messchaert M, Haer-Wigman L, Khan MI, Cremers FPM, Collin RWJ. EYS mutation update: in silico assessment of 271 reported and 26 novel variants in patients with retinitis pigmentosa. Hum Mutat. 2018;39(2):177-186. doi: 10.1002/humu.23371 [DOI] [PubMed] [Google Scholar]
  • 25.Cornelis SS, Bax NM, Zernant J, et al. In silico functional meta-analysis of 5,962 ABCA4 variants in 3,928 retinal dystrophy cases. Hum Mutat. 2017;38(4):400-408. doi: 10.1002/humu.23165 [DOI] [PubMed] [Google Scholar]
  • 26.Landrum MJ, Lee JM, Benson M, et al. ClinVar: improving access to variant interpretations and supporting evidence. Nucleic Acids Res. 2018;46(D1):D1062-D1067. doi: 10.1093/nar/gkx1153 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.ClinVar database. National Center for Biotechnology Information, US National Library of Medicine. Accessed April 28, 2020. http://www.clinvar.com/
  • 28.Shah M, Shanks M, Packham E, et al. Next generation sequencing using phenotype-based panels for genetic testing in inherited retinal diseases. Ophthalmic Genet. 2020;41(4):331-337. doi: 10.1080/13816810.2020.1778736 [DOI] [PubMed] [Google Scholar]
  • 29.Richards S, Aziz N, Bale S, et al. ; ACMG Laboratory Quality Assurance Committee . Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405-424. doi: 10.1038/gim.2015.30 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Levy JM, Yeh W-H, Pendse N, et al. Cytosine and adenine base editing of the brain, liver, retina, heart and skeletal muscle of mice via adeno-associated viruses. Nat Biomed Eng. 2020;4(1):97-110. doi: 10.1038/s41551-019-0501-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Suh S, Choi EH, Leinonen H, et al. Restoration of visual function in adult mice with an inherited retinal disease via adenine base editing. Nat Biomed Eng. Published online October 19, 2020. doi: 10.1038/s41551-020-00632-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Villiger L, Grisch-Chan HM, Lindsay H, et al. Treatment of a metabolic liver disease by in vivo genome base editing in adult mice. Nat Med. 2018;24(10):1519-1525. doi: 10.1038/s41591-018-0209-1 [DOI] [PubMed] [Google Scholar]
  • 33.Ryu S-M, Koo T, Kim K, et al. Adenine base editing in mouse embryos and an adult mouse model of Duchenne muscular dystrophy. Nat Biotechnol. 2018;36(6):536-539. doi: 10.1038/nbt.4148 [DOI] [PubMed] [Google Scholar]
  • 34.Jiang T, Henderson JM, Coote K, et al. Chemical modifications of adenine base editor mRNA and guide RNA expand its application scope. Nat Commun. 2020;11(1):1979. doi: 10.1038/s41467-020-15892-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Song C-Q, Jiang T, Richter M, et al. Adenine base editing in an adult mouse model of tyrosinaemia. Nat Biomed Eng. 2020;4(1):125-130. doi: 10.1038/s41551-019-0357-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Yeh W-H, Chiang H, Rees HA, Edge ASB, Liu DR. In vivo base editing of post-mitotic sensory cells. Nat Commun. 2018;9(1):2184. doi: 10.1038/s41467-018-04580-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Rees HA, Komor AC, Yeh W-H, et al. Improving the DNA specificity and applicability of base editing through protein engineering and protein delivery. Nat Commun. 2017;8:15790. doi: 10.1038/ncomms15790 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Sinnamon JR, Kim SY, Fisk JR, et al. In vivo repair of a protein underlying a neurological disorder by programmable RNA editing. Cell Rep. 2020;32(2):107878. doi: 10.1016/j.celrep.2020.107878 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Montiel-Gonzalez MF, Vallecillo-Viejo IC, Rosenthal JJC. An efficient system for selectively altering genetic information within mRNAs. Nucleic Acids Res. 2016;44(21):e157. doi: 10.1093/nar/gkw738 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Merkle T, Merz S, Reautschnig P, et al. Precise RNA editing by recruiting endogenous ADARs with antisense oligonucleotides. Nat Biotechnol. 2019;37(2):133-138. doi: 10.1038/s41587-019-0013-6 [DOI] [PubMed] [Google Scholar]
  • 41.Walton RT, Christie KA, Whittaker MN, Kleinstiver BP. Unconstrained genome targeting with near-PAMless engineered CRISPR-Cas9 variants. Science. 2020;368(6488):290-296. doi: 10.1126/science.aba8853 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Li X, Wang Y, Liu Y, et al. Base editing with a Cpf1-cytidine deaminase fusion. Nat Biotechnol. 2018;36(4):324-327. doi: 10.1038/nbt.4102 [DOI] [PubMed] [Google Scholar]
  • 43.Hu JH, Miller SM, Geurts MH, et al. Evolved Cas9 variants with broad PAM compatibility and high DNA specificity. Nature. 2018;556(7699):57-63. doi: 10.1038/nature26155 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Gaudelli NM, Lam DK, Rees HA, et al. Directed evolution of adenine base editors with increased activity and therapeutic application. Nat Biotechnol. 2020;38(7):892-900. doi: 10.1038/s41587-020-0491-6 [DOI] [PubMed] [Google Scholar]
  • 45.Gaudelli NM, Komor AC, Rees HA, et al. Programmable base editing of A•T to G•C in genomic DNA without DNA cleavage. Nature. 2017;551(7681):464-471. doi: 10.1038/nature24644 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Komor AC, Zhao KT, Packer MS, et al. Improved base excision repair inhibition and bacteriophage Mu Gam protein yields C:G-to-T:A base editors with higher efficiency and product purity. Sci Adv. 2017;3(8):eaao4774. doi: 10.1126/sciadv.aao4774 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Grunewald J, Zhou R, Iyer S, et al. CRISPR DNA base editors with reduced RNA off-target and self-editing activities. Nat Biotechnol. 2019;37(9):1041-1048. doi: 10.1038/s41587-019-0236-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Vallecillo-Viejo IC, Liscovitch-Brauer N, Montiel-Gonzalez MF, Eisenberg E, Rosenthal JJC. Abundant off-target edits from site-directed RNA editing can be reduced by nuclear localization of the editing enzyme. RNA Biol. 2018;15(1):104-114. doi: 10.1080/15476286.2017.1387711 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Vogel P, Moschref M, Li Q, et al. Efficient and precise editing of endogenous transcripts with SNAP-tagged ADARs. Nat Methods. 2018;15(7):535-538. doi: 10.1038/s41592-018-0017-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Stoltzfus A, Norris RW. On the causes of evolutionary transition:transversion bias. Mol Biol Evol. 2016;33(3):595-602. doi: 10.1093/molbev/msv274 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Kreutzer DA, Essigmann JM. Oxidized, deaminated cytosines are a source of C –> T transitions in vivo. Proc Natl Acad Sci U S A. 1998;95(7):3578-3582. doi: 10.1073/pnas.95.7.3578 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Lindahl T. Instability and decay of the primary structure of DNA. Nature. 1993;362(6422):709-715. doi: 10.1038/362709a0 [DOI] [PubMed] [Google Scholar]
  • 53.Cremers FPM, Lee W, Collin RWJ, Allikmets R. Clinical spectrum, genetic complexity and therapeutic approaches for retinal disease caused by ABCA4 mutations. Prog Retin Eye Res. 2020;79:100861. doi: 10.1016/j.preteyeres.2020.100861 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Sangermano R, Bax NM, Bauwens M, et al. Photoreceptor progenitor mRNA analysis reveals exon skipping resulting from the ABCA4 c.5461-10T→C mutation in Stargardt disease. Ophthalmology. 2016;123(6):1375-1385. doi: 10.1016/j.ophtha.2016.01.053 [DOI] [PubMed] [Google Scholar]
  • 55.Dreyer B, Tranebjaerg L, Brox V, et al. A common ancestral origin of the frequent and widespread 2299delG USH2A mutation. Am J Hum Genet. 2001;69(1):228-234. doi: 10.1086/321269 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Baux D, Blanchet C, Hamel C, et al. Enrichment of LOVD-USH bases with 152 USH2A genotypes defines an extensive mutational spectrum and highlights missense hotspots. Hum Mutat. 2014;35(10):1179-1186. doi: 10.1002/humu.22608 [DOI] [PubMed] [Google Scholar]
  • 57.Perez-Carro R, Blanco-Kelly F, Galbis-Martinez L, et al. Unravelling the pathogenic role and genotype-phenotype correlation of the USH2A p.(Cys759Phe) variant among Spanish families. PLoS One. 2018;13(6):e0199048. doi: 10.1371/journal.pone.0199048 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Slijkerman RW, Vache C, Dona M, et al. Antisense oligonucleotide-based splice correction for USH2A-associated retinal degeneration caused by a frequent deep-intronic mutation. Mol Ther Nucleic Acids. 2016;5(10):e381. doi: 10.1038/mtna.2016.89 [DOI] [PubMed] [Google Scholar]
  • 59.ProQR Therapeutics . Study to evaluate safety and tolerability of QR-421a in subjects with RP due to mutations in exon 13 of the USH2A gene (stellar). ClinicalTrials.gov identifier: NCT03780257. Accessed May 5, 2020. https://clinicaltrials.gov/ct2/show/NCT03780257
  • 60.Gapinske M, Luu A, Winter J, et al. CRISPR-SKIP: programmable gene splicing with single base editors. Genome Biol. 2018;19(1):107. doi: 10.1186/s13059-018-1482-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Konermann S, Lotfy P, Brideau NJ, Oki J, Shokhirev MN, Hsu PD. Transcriptome engineering with RNA-targeting type VI-D CRISPR effectors. Cell. 2018;173(3):665-676. doi: 10.1016/j.cell.2018.02.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Bonnet C, Riahi Z, Chantot-Bastaraud S, et al. An innovative strategy for the molecular diagnosis of Usher syndrome identifies causal biallelic mutations in 93% of European patients. Eur J Hum Genet. 2016;24(12):1730-1738. doi: 10.1038/ejhg.2016.99 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63.Nagel-Wolfrum K, Moller F, Penner I, Wolfrum U. Translational read-through as an alternative approach for ocular gene therapy of retinal dystrophies caused by in-frame nonsense mutations. Vis Neurosci. 2014;31(4-5):309-316. doi: 10.1017/S0952523814000194 [DOI] [PubMed] [Google Scholar]
  • 64.Sheck L, Davies WIL, Moradi P, et al. Leber congenital amaurosis associated with mutations in CEP290, clinical phenotype, and natural history in preparation for trials of novel therapies. Ophthalmology. 2018;125(6):894-903. doi: 10.1016/j.ophtha.2017.12.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65.Dulla K, Aguila M, Lane A, et al. Splice-modulating oligonucleotide QR-110 restores CEP290 mRNA and function in human c.2991+1655A>G LCA10 models. Mol Ther Nucleic Acids. 2018;12:730-740. doi: 10.1016/j.omtn.2018.07.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Cideciyan AV, Jacobson SG, Drack AV, et al. Effect of an intravitreal antisense oligonucleotide on vision in Leber congenital amaurosis due to a photoreceptor cilium defect. Nat Med. 2019;25(2):225-228. doi: 10.1038/s41591-018-0295-0 [DOI] [PubMed] [Google Scholar]
  • 67.Jacobson SG, Cideciyan AV, Gibbs D, et al. Retinal disease course in Usher syndrome 1B due to MYO7A mutations. Invest Ophthalmol Vis Sci. 2011;52(11):7924-7936. doi: 10.1167/iovs.11-8313 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68.Astuto LM, Bork JM, Weston MD, et al. CDH23 mutation and phenotype heterogeneity: a profile of 107 diverse families with Usher syndrome and nonsyndromic deafness. Am J Hum Genet. 2002;71(2):262-275. doi: 10.1086/341558 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69.Mizutari K, Mutai H, Namba K, et al. High prevalence of CDH23 mutations in patients with congenital high-frequency sporadic or recessively inherited hearing loss. Orphanet J Rare Dis. 2015;10:60. doi: 10.1186/s13023-015-0276-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70.Bork JM, Peters LM, Riazuddin S, et al. Usher syndrome 1D and nonsyndromic autosomal recessive deafness DFNB12 are caused by allelic mutations of the novel cadherin-like gene CDH23. Am J Hum Genet. 2001;68(1):26-37. doi: 10.1086/316954 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71.Valero R, de Castro-Miro M, Jimenez-Ochoa S, Rodriguez-Ezcurra JJ, Marfany G, Gonzalez-Duarte R. Aberrant splicing events associated to CDH23 noncanonical splice site mutations in a proband with atypical Usher syndrome 1. Genes (Basel). 2019;10(10):732. doi: 10.3390/genes10100732 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 72.Quinn J, Musa A, Kantor A, et al. Genome-editing strategies for treating human retinal degenerations. Hum Gene Ther. Published online November 19, 2020. doi: 10.1089/hum.2020.231 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 73.Xue K, MacLaren RE. Antisense oligonucleotide therapeutics in clinical trials for the treatment of inherited retinal diseases. Expert Opin Investig Drugs. 2020;29(10):1163-1170. doi: 10.1080/13543784.2020.1804853 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement.

eTable 1. Pathogenic Alleles From the LOVD Broken Down by Variation Type and Gene, as Analyzed in Figure 1

eTable 2. Most Common Variants in the LOVD

eTable 3. Pathogenic Alleles From the Oxford Database Broken Down by Variation Type and Gene, as Analyzed in Figure 3A

eTable 4. Summary of the 5 Most Common Alleles for Genes ABCA4 and USH2A in the Oxford Data Set

eTable 5. Summary of the 5 Most Common Alleles for Genes ABCA4 and USH2A in the Stone et al17 Data Set

eFigure. Flow Diagram Outlining Method of LOVD Analysis

Click here for additional data file. (442.6KB, pdf)

Articles from JAMA Ophthalmology are provided here courtesy of American Medical Association

RESOURCES