Species-dependent splice recognition of a cryptic exon resulting from a recurrent intronic CEP290 mutation that causes congenital blindness

Retinal Ciliopathies and Potential Gene Therapies: A Focus on Human iPSC-Derived Organoid Models

The human photoreceptor function is dependent on a highly specialised cilium. Perturbation of cilial function can often lead to death of the photoreceptor and loss of vision. Retinal ciliopathies are a genetically diverse range of inherited retinal disorders affecting aspects of the photoreceptor cilium. Despite advances in the understanding of retinal ciliopathies utilising animal disease models, they can often lack the ability to accurately mimic the observed patient phenotype, possibly due to structural and functional deviations from the human retina. Human-induced pluripotent stem cells (hiPSCs) can be utilised to generate an alternative disease model, the 3D retinal organoid, which contains all major retinal cell types including photoreceptors complete with cilial structures. These retinal organoids facilitate the study of disease mechanisms and potential therapies in a human-derived system. Three-dimensional retinal organoids are still a developing technology, and despite impressive progress, several limitations remain. This review will discuss the state of hiPSC-derived retinal organoid technology for accurately modelling prominent retinal ciliopathies related to genes, including RPGR, CEP290, MYO7A, and USH2A. Additionally, we will discuss the development of novel gene therapy approaches targeting retinal ciliopathies, including the delivery of large genes and gene-editing techniques.

Rational design of a genomically humanized mouse model for dominantly inherited hearing loss, DFNA9

DFNA9 is a dominantly inherited form of adult-onset progressive hearing impairment caused by mutations in the COCH gene. COCH encodes cochlin, a crucial extracellular matrix protein. We established a genomically humanized mouse model for the Dutch/Belgian c.151C>T founder mutation in COCH. Considering upcoming sequence-specific genetic therapies, we exchanged the genomic murine Coch exons 3-6 for the corresponding human sequence. Introducing human-specific genetic information into mouse exons can be risky. To mitigate unforeseen consequences on cochlin function resulting from the introduction of the human COCH protein-coding sequence, we converted all human-specific amino acids to mouse equivalents. We furthermore optimized the recognition of the human COCH exons by the murine splicing machinery during pre-mRNA splicing. Subsequent observations in mouse embryonic stem cells revealed correct splicing of the hybrid Coch transcript. The inner ear of the established humanized Coch mice displays correctly-spliced wild-type and mutant humanized Coch alleles. For a comprehensive study of auditory function, mice were crossbred with C57BL/6 Cdh23753A>G mice to remove the Cdh23ahl allele from the genetic background of the mice. At 9 months, all humanized Coch genotypes showed hearing thresholds comparable to wild-type C57BL/6 Cdh23753A>G mice. This indicates that both the introduction of human wildtype COCH, and correction of Cdh23ahl in the humanized Coch lines was successful. Overall, our approach proved beneficial in eliminating potential adverse events of genomic humanization of mouse genes, and provides us with a model in which sequence-specific therapies directed against the human mutant COCH alle can be investigated. With the hearing and balance defects anticipated to occur late in the second year of life, a long-term follow-up study is ongoing to fully characterize the humanized Coch mouse model.

Utilization of the retinal organoid model to evaluate the feasibility of genetic strategies to ameliorate retinal disease(s)

Organoid models have quickly become a popular research tool to evaluate novel therapeutics on 3-D recapitulated tissue. This has enabled researchers to use physiologically relevant human tissue in vitro to augment the standard use of immortalized cells and animal models. Organoids can also provide a model when an engineered animal cannot recreate a specific disease phenotype. In particular, the retinal research field has taken advantage of this burgeoning technology to provide insight into inherited retinal disease(s) mechanisms and therapeutic intervention to ameliorate their effects. In this review we will discuss the use of both wild-type and patient-specific retinal organoids to further gene therapy research that could potentially prevent retinal disease(s) progression. Furthermore, we will discuss the pitfalls of current retinal organoid technology and present potential solutions that could overcome these hurdles in the near future.

Pathogenic Variants in CEP290 or IQCB1 Cause Earlier-Onset Retinopathy in Senior-Loken Syndrome Compared to Those in INVS, NPHP3, or NPHP4

PURPOSE

Senior-Loken syndrome (SLSN) is an autosomal recessive disorder characterized by retinopathy and nephronophthisis. This study aimed to evaluate whether different phenotypes are associated with different variants or subsets of 10 SLSN-associated genes based on an in-house data set and a literature review.

DESIGN

Retrospective case series.

METHODS

Patients with biallelic variants in SLSN-associated genes, including NPHP1, INVS, NPHP3, NPHP4, IQCB1, CEP290, SDCCAG8, WDR19, CEP164, and TRAF3IP1, were recruited. Ocular phenotypes and nephrology medical records were collected for comprehensive analysis.

RESULTS

Variants in 5 genes were identified in 74 patients from 70 unrelated families, including CEP290 (61.4%), IQCB1 (28.6%), NPHP1 (4.2%), NPHP4 (2.9%), and WDR19 (2.9%). The median age at the onset of retinopathy was approximately 1 month (since birth). Nystagmus was the most common initial sign in patients with CEP290 (28 of 44, 63.6%) or IQCB1 (19 of 22, 86.4%) variants. Cone and rod responses were extinguished in 53 of 55 patients (96.4%). Characteristic fundus changes were observed in CEP290- and IQCB1-associated patients. During follow-up, 70 of the 74 patients were referred to nephrology, among whom nephronophthisis was not detected in 62 patients (88.6%) at a median age of 6 years but presented in 8 patients (11.4%) aged approximately 9 years.

CONCLUSIONS

Patients with pathogenic variants in CEP290 or IQCB1 presented early with retinopathy, whereas other patients with INVS, NPHP3, or NPHP4 variants first developed nephropathy. Therefore, awareness of the genetic and clinical features may facilitate the clinical management of SLSN, especially early intervention of kidney problems for patients with eyes affected first.

Experimental Model Systems Used in the Preclinical Development of Nucleic Acid Therapeutics

Preclinical evaluation of nucleic acid therapeutics (NATs) in relevant experimental model systems is essential for NAT drug development. As part of COST Action "DARTER" (Delivery of Antisense RNA ThERapeutics), a network of researchers in the field of RNA therapeutics, we have conducted a survey on the experimental model systems routinely used by our members in preclinical NAT development. The questionnaire focused on both cellular and animal models. Our survey results suggest that skin fibroblast cultures derived from patients is the most commonly used cellular model, while induced pluripotent stem cell-derived models are also highly reported, highlighting the increasing potential of this technology. Splice-switching antisense oligonucleotide is the most frequently investigated RNA molecule, followed by small interfering RNA. Animal models are less prevalent but also widely used among groups in the network, with transgenic mouse models ranking the top. Concerning the research fields represented in our survey, the mostly studied disease area is neuromuscular disorders, followed by neurometabolic diseases and cancers. Brain, skeletal muscle, heart, and liver are the top four tissues of interest reported. We expect that this snapshot of the current preclinical models will facilitate decision making and the share of resources between academics and industry worldwide to facilitate the development of NATs.

The Predicted Splicing Variant c.11+5G>A in RPE65 Leads to a Reduction in mRNA Expression in a Cell-Specific Manner

Pathogenic variants in RPE65 lead to retinal diseases, causing a vision impairment. In this work, we investigated the pathomechanism behind the frequent RPE65 variant, c.11+5G>A. Previous in silico predictions classified this change as a splice variant. Our prediction using novel software's suggested a 124-nt exon elongation containing a premature stop codon. This elongation was validated using midigenes-based approaches. Similar results were observed in patient-derived induced pluripotent stem cells (iPSC) and photoreceptor precursor cells. However, the splicing defect in all cases was detected at low levels and thereby does not fully explain the recessive condition of the resulting disease. Long-read sequencing discarded other rearrangements or variants that could explain the diseases. Subsequently, a more relevant model was employed: iPSC-derived retinal pigment epithelium (RPE) cells. In patient-derived iPSC-RPE cells, the expression of RPE65 was strongly reduced even after inhibiting a nonsense-mediated decay, contradicting the predicted splicing defect. Additional experiments demonstrated a cell-specific gene expression reduction due to the presence of the c.11+5G>A variant. This decrease also leads to the lack of the RPE65 protein, and differences in size and pigmentation between the patient and control iPSC-RPE. Altogether, our data suggest that the c.11+5G>A variant causes a cell-specific defect in the expression of RPE65 rather than the anticipated splicing defect which was predicted in silico.

Reproducibility, relevance and reliability as barriers to efficient and credible biomedical technology translation

Biomedical research accuracy and relevance for improving healthcare are increasingly identified as costly problems. Basic research data quality, reporting and methodology, and reproducibility are common factors implicated in this challenge. Preclinical models of disease and therapy, largely conducted in rodents, have known deficiencies in replicating most human conditions. Their translation to human results is acknowledged to be poor for decades. Clinical data quality and quantity is also recognized as deficient; gold standard randomized clinical trials are expensive. Few solid conclusions from clinical studies are replicable and many remain unpublished. The translational pathway from fundamental biomedical research through to innovative solutions handed to clinical practitioners is therefore highly inefficient and costly in terms of wasted resources, early claims from fundamental discoveries never witnessed in humans, and few new, improved solutions available clinically for myriad diseases. Improving this biomedical research strategy and resourcing for reliability, translational relevance, reproducibility and clinical impact requires careful analysis and consistent enforcement at both funding and peer review levels.

Considerations for Generating Humanized Mouse Models to Test Efficacy of Antisense Oligonucleotides

Over the last decades, animal models have become increasingly important in therapeutic drug development and assessment. The use of these models, mainly mice and rats, allow evaluating drugs in the real-organism environment and context. However, several molecular therapeutic approaches are sequence-dependent, and therefore, the humanization of such models is required to assess the efficacy. The generation of genetically modified humanized mouse models is often an expensive and laborious process that may not always recapitulate the human molecular and/or physiological phenotype. In this chapter, we summarize basic aspects to consider before designing and generating humanized models, especially when they are aimed to test antisense-based therapies.

Exon Skipping Through Chimeric Antisense U1 snRNAs to Correct Retinitis Pigmentosa GTPase-Regulator (RPGR) Splice Defect

Inherited retinal dystrophies are caused by mutations in more than 250 genes, each of them carrying several types of mutations that can lead to different clinical phenotypes. Mutations in Retinitis Pigmentosa GTPase-Regulator (RPGR) cause X-linked Retinitis pigmentosa (RP). A nucleotide substitution in intron 9 of RPGR causes the increase of an alternatively spliced isoform of the mature mRNA, bearing exon 9a (E9a). This introduces a stop codon, leading to truncation of the protein. Aiming at restoring impaired gene expression, we developed an antisense RNA-based therapeutic approach for the skipping of RPGR E9a. We designed a set of specific U1 antisense snRNAs (U1_asRNAs) and tested their efficacy in vitro, upon transient cotransfection with RPGR minigene reporter systems in HEK-293T, 661W, and PC-12 cell lines.

Delivery of oligonucleotide-based therapeutics: challenges and opportunities

Nucleic acid-based therapeutics that regulate gene expression have been developed towards clinical use at a steady pace for several decades, but in recent years the field has been accelerating. To date, there are 11 marketed products based on antisense oligonucleotides, aptamers and small interfering RNAs, and many others are in the pipeline for both academia and industry. A major technology trigger for this development has been progress in oligonucleotide chemistry to improve the drug properties and reduce cost of goods, but the main hurdle for the application to a wider range of disorders is delivery to target tissues. The adoption of delivery technologies, such as conjugates or nanoparticles, has been a game changer for many therapeutic indications, but many others are still awaiting their eureka moment. Here, we cover the variety of methods developed to deliver nucleic acid-based therapeutics across biological barriers and the model systems used to test them. We discuss important safety considerations and regulatory requirements for synthetic oligonucleotide chemistries and the hurdles for translating laboratory breakthroughs to the clinic. Recent advances in the delivery of nucleic acid-based therapeutics and in the development of model systems, as well as safety considerations and regulatory requirements for synthetic oligonucleotide chemistries are discussed in this review on oligonucleotide-based therapeutics.

Improving Translation by Identifying Evidence for More Human-Relevant Preclinical Strategies

Preclinical animal studies are performed to analyse the safety and efficacy of new treatments, with the aim to protect humans. However, there are questions and concerns about the quality and usefulness of preclinical animal research. Translational success rates vary between 0 and 100%, and no clear relationship has been found with possible predictive factors such as animal species or field of research. Therefore, it is not yet possible to indicate what factors predict successful translation. Translational strategies were therefore discussed at an international conference held in the Netherlands in November 2019, aiming to develop practical guidelines for more robust animal-to-human translation. The conference was organised during the course of a research project funded by the Dutch Research Council (313-99-310), addressing possible solutions for the low translational values that had been published for a multitude of animal studies in human health care. This article provides an overview of the project and the conference discussions. Based on the conference results and the findings from the research project, we define four points of attention that are crucial in the search for improved translational success rates: (a) optimising the methods and design of studies; (b) incorporation of the complexity of the human patient in research; (c) start with the patient rather than existing animal models as the gold standard; and (d) more and better collaboration within the chain from funding to pharmacy. We conclude that this requires improved organization and use of procedures, as well as a change of attitude and culture in research, including a consideration of the translational value of animal-free innovations and human-relevant science.

The Scope for Thalassemia Gene Therapy by Disruption of Aberrant Regulatory Elements

The common IVSI-110 (G>A) β-thalassemia mutation is a paradigm for intronic disease-causing mutations and their functional repair by non-homologous end joining-mediated disruption. Such mutation-specific repair by disruption of aberrant regulatory elements (DARE) is highly efficient, but to date, no systematic analysis has been performed to evaluate disease-causing mutations as therapeutic targets. Here, DARE was performed in highly characterized erythroid IVSI-110(G>A) transgenic cells and the disruption events were compared with published observations in primary CD34⁺ cells. DARE achieved the functional correction of β-globin expression equally through the removal of causative mutations and through the removal of context sequences, with disruption events and the restriction of indel events close to the cut site closely resembling those seen in primary cells. Correlation of DNA-, RNA-, and protein-level findings then allowed the extrapolation of findings to other mutations by in silico analyses for potential repair based on the clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated (Cas) 9, Cas12a, and transcription activator-like effector nuclease (TALEN) platforms. The high efficiency of DARE and unexpected freedom of target design render the approach potentially suitable for 14 known thalassemia mutations besides IVSI-110(G>A) and put it forward for several prominent mutations causing other inherited diseases. The application of DARE, therefore, has a wide scope for sustainable personalized advanced therapy medicinal product development for thalassemia and beyond.

The use of genetically humanized animal models for personalized medicine approaches

For many genetic diseases, researchers are developing personalized medicine approaches. These sometimes employ custom genetic interventions such as antisense-mediated exon skipping or genome editing, aiming to restore protein function in a mutation-specific manner. Animal models can facilitate the development of personalized medicine approaches; however, given that they target human mutations and therefore human genetic sequences, scientists rely on the availability of humanized animal models. Here, we outline the usefulness, caveats and potential of such models, using the example of the hDMDdel52/mdx model, a humanized model recently generated for Duchenne muscular dystrophy (DMD).

ATP-binding cassette subfamily A, member 4 intronic variants c.4773+3A>G and c.5461-10T>C cause Stargardt disease due to defective splicing

PURPOSE

Inherited retinal dystrophies (IRDs) represent a group of progressive conditions affecting the retina. There is a great genetic heterogeneity causing IRDs, and to date, more than 260 genes are associated with IRDs. Stargardt disease, type 1 (STGD1) or macular degeneration with flecks, STGD1 represents a disease with early onset, central visual impairment, frequent appearance of yellowish flecks and mutations in the ATP-binding cassette subfamily A, member 4 (ABCA4) gene. A large number of intronic sequence variants in ABCA4 have been considered pathogenic although their functional effect was seldom demonstrated. In this study, we aimed to reveal how intronic variants present in patients with Stargardt from the same Swedish family affect splicing.

METHODS

The splicing of the ABCA4 gene was studied in human embryonic kidney cells, HEK293T, and in human retinal pigment epithelium cells, ARPE-19, using a minigene system containing variants c.4773+3A>G and c.5461-10T>C.

RESULTS

We showed that both ABCA4 variants, c.4773+3A>G and c.5461-10T>C, cause aberrant splicing of the ABCA4 minigene resulting in exon skipping. We also demonstrated that splicing of ABCA4 has different outcomes depending on transfected cell type.

CONCLUSION

Two intronic variants c.4773+3A>G and c.5461-10T>C, both predicted to affect splicing, are indeed disease-causing mutations due to skipping of exons 33, 34, 39 and 40 of ABCA4 gene. The experimental proof that ABCA4 mutations in STGD patients affect protein function is crucial for their inclusion to future clinical trials; therefore, functional testing of all ABCA4 intronic variants associated with Stargardt disease by minigene technology is desirable.

Poor Splice-Site Recognition in a Humanized Zebrafish Knockin Model for the Recurrent Deep-Intronic c.7595-2144A>G Mutation in USH2A

The frequent deep-intronic c.7595-2144A>G mutation in intron 40 of USH2A generates a high-quality splice donor site, resulting in the incorporation of a pseudoexon (PE40) into the mature transcript that is predicted to prematurely terminate usherin translation. Aberrant USH2A pre-mRNA splicing could be corrected in patient-derived fibroblasts using antisense oligonucleotides. With the aim to study the effect of the c.7595-2144A>G mutation and USH2A splice redirection on retinal function, a humanized zebrafish knockin model was generated, in which 670 basepairs of ush2a intron 40 were exchanged for 557 basepairs of the corresponding human sequence using an optimized CRISPR/Cas9-based protocol. However, in the retina of adult homozygous humanized zebrafish, only 7.4% ± 3.9% of ush2a transcripts contained the human PE40 sequence and immunohistochemical analyses revealed no differences in the usherin expression and localization between the retina of humanized and wild-type zebrafish larvae. Nevertheless, we were able to partially correct aberrant ush2a splicing using a PE40-targeting antisense morpholino. Our results indicate a clear difference in splice-site recognition by the human and zebrafish splicing machinery. Therefore, we propose a protocol in which the effect of human splice-modulating mutations is studied in a zebrafish-specific cell-based splice assay before the generation of a humanized zebrafish knockin model.

Splice-Modulating Oligonucleotide QR-110 Restores CEP290 mRNA and Function in Human c.2991+1655A>G LCA10 Models

Leber congenital amaurosis type 10 (LCA10) is a severe inherited retinal dystrophy associated with mutations in CEP290. The deep intronic c.2991+1655A>G mutation in CEP290 is the most common mutation in LCA10 individuals and represents an ideal target for oligonucleotide therapeutics. Here, a panel of antisense oligonucleotides was designed to correct the splicing defect associated with the mutation and screened for efficacy and safety. This identified QR-110 as the best-performing molecule. QR-110 restored wild-type CEP290 mRNA and protein expression levels in CEP290 c.2991+1655A>G homozygous and compound heterozygous LCA10 primary fibroblasts. Furthermore, in homozygous three-dimensional iPSC-derived retinal organoids, QR-110 showed a dose-dependent restoration of mRNA and protein function, as measured by percentage and length of photoreceptor cilia, without off-target effects. Localization studies in wild-type mice and rabbits showed that QR-110 readily reached all retinal layers, with an estimated half-life of 58 days. It was well tolerated following intravitreal injection in monkeys. In conclusion, the pharmacodynamic, pharmacokinetic, and safety properties make QR-110 a promising candidate for treating LCA10, and clinical development is currently ongoing.

Design and In Vitro Use of Antisense Oligonucleotides to Correct Pre-mRNA Splicing Defects in Inherited Retinal Dystrophies

Antisense oligonucleotides (AONs) are small molecules able to bind to the pre-mRNA and modulate splicing. The increasing amount of intronic mutations leading to pseudoexon insertion in genes underlying inherited retinal dystrophies (IRDs) has highlighted the potential of AONs as a therapeutic tool for these disorders. Here we describe how to design and test AON molecules in vitro in order to correct pre-mRNA splicing defects involved in IRDs.

Delivery is key: lessons learnt from developing splice-switching antisense therapies

The use of splice‐switching antisense therapy is highly promising, with a wealth of pre‐clinical data and numerous clinical trials ongoing. Nevertheless, its potential to treat a variety of disorders has yet to be realized. The main obstacle impeding the clinical translation of this approach is the relatively poor delivery of antisense oligonucleotides to target tissues after systemic delivery. We are a group of researchers closely involved in the development of these therapies and would like to communicate our discussions concerning the validity of standard methodologies currently used in their pre‐clinical development, the gaps in current knowledge and the pertinent challenges facing the field. We therefore make recommendations in order to focus future research efforts and facilitate a wider application of therapeutic antisense oligonucleotides.

Leveraging splice-affecting variant predictors and a minigene validation system to identify Mendelian disease-causing variants among exon-captured variants of uncertain significance

The genetic heterogeneity of Mendelian disorders results in a significant proportion of patients that are unable to be assigned a confident molecular diagnosis after conventional exon sequencing and variant interpretation. Here, we evaluated how many patients with an inherited retinal disease (IRD) have variants of uncertain significance (VUS) that are disrupting splicing in a known IRD gene by means other than affecting the canonical dinucleotide splice site. Three in silico splice-affecting variant predictors were leveraged to annotate and prioritize variants for splicing functional validation. An in vitro minigene system was used to assay each variant's effect on splicing. Starting with 745 IRD patients lacking a confident molecular diagnosis, we validated 23 VUS as splicing variants that likely explain disease in 26 patients. Using our results, we optimized in silico score cutoffs to guide future variant interpretation. Variants that alter base pairs other than the canonical GT-AG dinucleotide are often not considered for their potential effect on RNA splicing but in silico tools and a minigene system can be utilized for the prioritization and validation of such splice-disrupting variants. These variants can be overlooked causes of human disease but can be identified using conventional exon sequencing with proper interpretation guidelines.

Retinas in a Dish Peek into Inherited Retinal Degeneration

Human retinal degeneration can cause blindness, and the lack of relevant model systems has made identifying underlying mechanisms challenging. Parfitt et al. (2016) generate three-dimensional retinal tissue from patient-derived induced pluripotent stem cells to identify how CEP290 mutations cause retinal degeneration, and show an antisense approach can correct disease-associated phenotypes.

Antisense Oligonucleotide-based Splice Correction for USH2A-associated Retinal Degeneration Caused by a Frequent Deep-intronic Mutation

Usher syndrome (USH) is the most common cause of combined deaf-blindness in man. The hearing loss can be partly compensated by providing patients with hearing aids or cochlear implants, but the loss of vision is currently untreatable. In general, mutations in the USH2A gene are the most frequent cause of USH explaining up to 50% of all patients worldwide. The first deep-intronic mutation in the USH2A gene (c.7595-2144A>G) was reported in 2012, leading to the insertion of a pseudoexon (PE40) into the mature USH2A transcript. When translated, this PE40-containing transcript is predicted to result in a truncated non-functional USH2A protein. In this study, we explored the potential of antisense oligonucleotides (AONs) to prevent aberrant splicing of USH2A pre-mRNA as a consequence of the c.7595-2144A>G mutation. Engineered 2'-O-methylphosphorothioate AONs targeting the PE40 splice acceptor site and/or exonic splice enhancer regions displayed significant splice correction potential in both patient derived fibroblasts and a minigene splice assay for USH2A c.7595-2144A>G, whereas a non-binding sense oligonucleotide had no effect on splicing. Altogether, AON-based splice correction could be a promising approach for the development of a future treatment for USH2A-associated retinitis pigmentosa caused by the deep-intronic c.7595-2144A>G mutation.

Photoreceptor Progenitor mRNA Analysis Reveals Exon Skipping Resulting from the ABCA4 c.5461-10T→C Mutation in Stargardt Disease

PURPOSE

To elucidate the functional effect of the ABCA4 variant c.5461-10T→C, one of the most frequent variants associated with Stargardt disease (STGD1).

DESIGN

Case series.

PARTICIPANTS

Seventeen persons with STGD1 carrying ABCA4 variants and 1 control participant.

METHODS

Haplotype analysis of 4 homozygotes and 11 heterozygotes for c.5461-10T→C and sequence analysis of the ABCA4 gene for a homozygous proband. Fibroblasts were reprogrammed from 3 persons with STGD1 into induced pluripotent stem cells, which were differentiated into photoreceptor progenitor cells (PPCs). The effect of the c.5461-10T→C variant on RNA splicing by reverse-transcription polymerase chain reaction was analyzed using PPC mRNA. In vitro assays were performed with minigene constructs containing ABCA4 exon 39. We analyzed the natural history and ophthalmologic characteristics of 4 persons homozygous for c.5461-10T→C.

MAIN OUTCOME MEASURES

Haplotype and rare variant data for ABCA4, RNA splice defects, age at diagnosis, visual acuity, fundus appearance, visual field, electroretinography (ERG) results, fluorescein angiography results, and fundus autofluorescence findings.

RESULTS

The frequent ABCA4 variant c.5461-10T→C has a subtle effect on splicing based on prediction programs. A founder haplotype containing c.5461-10T→C was found to span approximately 96 kb of ABCA4 and did not contain other rare sequence variants. Patient-derived PPCs showed skipping of exon 39 or exons 39 and 40 in the mRNA. HEK293T cell transduction with minigenes carrying exon 39 showed that the splice defects were the result of the c.5461-10T→C variant. All 4 subjects carrying the c.5461-10T→C variant in a homozygous state showed a young age of STGD1 onset, with low visual acuity at presentation and abnormal cone ERG results. All 4 demonstrated severe cone-rod dystrophy before 20 years of age and were legally blind by 25 years of age.

CONCLUSIONS

The ABCA4 variant c.5461-10T→C is located on a founder haplotype lacking other disease-causing rare sequence variants. In vitro studies revealed that it leads to mRNA exon skipping and ABCA4 protein truncation. Given the severe phenotype in persons homozygous for this variant, we conclude that this variant results in the absence of ABCA4 activity.