Retinitis pigmentosa (RP) is a disease that primarily affects the peripheral retina and ultimately causes visual impairment. X-chromosomal forms of RP are frequently caused by mutations in the retinitis pigmentosa GTPase regulator (RPGR) gene. We show that the novel splice donor site (SDS) mutation c.1245+3A>T in intron 10 of RPGR cosegregates with RP in a five-generation Caucasian family. The mutation causes in-frame skipping of exon 10 from RPGR transcripts in patient-derived primary fibroblasts. To correct the splice defect, we developed a gene therapeutic approach using mutation-adapted U1 small nuclear RNA (U1). U1 is required for SDS recognition of pre-mRNAs and initiates the splice process. The mutation described herein interferes with the recognition of the SDS by U1. To overcome the deleterious effects of the mutation, we generated four U1 isoforms with increasing complementarity to the SDS. Lentiviral particles were used to transduce patient-derived fibroblasts with these U1 variants. Full complementarity of U1 corrects the splice defect partially and increases recognition of the mutant SDS. The therapeutic effect is U1-concentration dependent as we show for endogenously expressed RPGR transcripts in patient-derived cells. U1-based gene therapeutic approaches constitute promising technologies to treat SDS mutations in inherited diseases including X-linked RP.

Leber congenital amaurosis (LCA) is a rare inherited retinal disorder affecting approximately 1:50,000 people worldwide. So far, mutations in 25 genes have been associated with LCA, with CEP290 (encoding the Centrosomal protein of 290 kDa) being the most frequently mutated gene. The most recurrent LCA-causing CEP290 mutation, c.2991+1655A>G, causes the insertion of a pseudoexon into a variable proportion of CEP290 transcripts. We previously demonstrated that antisense oligonucleotides (AONs) have a high therapeutic potential for patients homozygously harbouring this mutation, although to date, it is unclear whether rescuing one single allele is enough to restore CEP290 function. Here, we assessed the AON efficacy at RNA, protein and cellular levels in samples that are compound heterozygous for this mutation, together with a protein-truncating mutation in CEP290. We demonstrate that AONs can efficiently restore splicing and increase protein levels. However, due to a high variability in ciliation among the patient-derived cell lines, the efficacy of the AONs was more difficult to assess at the cellular level. This observation points towards the importance of the severity of the second allele and possibly other genetic variants present in each individual. Overall, AONs seem to be a promising tool to treat CEP290-associated LCA, not only in homozygous but also in compound heterozygous carriers of the c.2991+1655A>G variant.

Purpose: Gene therapies to treat eye disorders have been extensively studied in the past 20 years. Frequently, adeno-associated viruses were applied to the subretinal or intravitreal space of the eye to transduce retinal cells with nucleotide sequences of therapeutic potential. In this study we describe a novel intravitreal injection procedure that leads to a reproducible adeno-associated virus (AAV)2/8-mediated transduction of more than 70% of the retina. Methods: Prior to a single intravitreal injection of a enhanced green fluorescent protien (GFP)-expressing viral suspension, we performed an aspiration of vitreous tissue from wild-type C57Bl/6J mice. One and one-half microliters of AAV2/8 suspension was injected. Funduscopy, optical coherence tomography (OCT), laser scanning microscopy of retinal flat mounts, cryosections of eye cups, and ERG recordings verified the efficacy and safety of the method. Results: The combination of vitreous aspiration and intravitreal injection resulted in an almost complete transduction of the retina in approximately 60% of the eyes and showed transduced cells in all retinal layers. Photoreceptors and RPE cells were predominantly transduced. Eyes presented with well-preserved retinal morphology. Electroretinographic recordings suggested that the new combination of techniques did not cause significant alterations of the retinal physiology. Conclusions: We show a novel application technique of AAV2/8 to the vitreous of mice that leads to widespread transduction of the retina. The results of this study have implications for virus-based gene therapies and basic science; for example, they might provide an approach to apply gene replacement strategies or clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 in vivo. It may further help to develop similar techniques for larger animal models or humans.

Manipulation of pre-mRNA processing is a promising approach toward overcoming disease-causing mutations and treating human diseases. We show that a combined treatment applying two splice-manipulating technologies improves therapeutic efficacies to correct mutation-induced splice defects. Previously, we identified a family affected by retinitis pigmentosa caused by the homozygous BBS1 splice donor site mutation c.479G > A. The mutation leads to both exon 5 skipping and intron 5 retention. We developed a therapeutic approach applying lentivirus-mediated gene delivery of engineered U1 small nuclear RNA (U1), which resulted in increased levels of correctly spliced BBS1. Herein, we show that the therapeutic effect of the engineered U1 efficiently reverted exon skipping but failed to reduce the intron retention. To complement the engineered U1 treatment, we identified four different antisense oligonucleotides (AONs) that block intron 5 retention in BBS1 transcripts. A treatment using engineered U1 in combination with AONs showed the highest therapeutic efficacy and increased the amount of correctly spliced BBS1 transcripts. We did not detect elevated levels of apoptotic cell death in AON-treated cell lines. In conclusion, engineered U1 or AONs provide efficient therapies with complementary effects and can be combined to increase efficacy of therapeutic approaches to correct splice defects.

Autosomal dominant optic atrophy (ADOA) is frequently caused by mutations in the optic atrophy 1 (OPA1) gene, with haploinsufficiency being the major genetic pathomechanism. Almost 30% of the OPA1-associated cases suffer from splice defects. We identified a novel OPA1 mutation, c.1065+5G>A, in patients with ADOA. In patient-derived fibroblasts, the mutation led to skipping of OPA1 exon 10, reducing the OPA1 protein expression by approximately 50%. We developed a molecular treatment to correct the splice defect in OPA1 using engineered U1 splice factors retargeted to different locations in OPA1 exon 10 or intron 10. The strongest therapeutic effect was detected when U1 binding was engineered to bind to intron 10 at position +18, a position predicted by bioinformatics to be a promising binding site. We were able to significantly silence the effect of the mutation (skipping of exon 10) and simultaneously increase the expression level of normal transcripts. Retargeting U1 to the canonical splice donor site did not lead to a detectable splice correction. This proof-of-concept study indicates for the first time the feasibility of splice mutation correction as a treatment option for ADOA. Increasing the amount of correctly spliced OPA1 transcripts may suffice to overcome the haploinsufficiency.