CRB1 is required for recycling by RAB11A+ vesicles in human retinal organoids.

CRB1 gene mutations can cause early- or late-onset retinitis pigmentosa, Leber congenital amaurosis, or maculopathy. Recapitulating human CRB1 phenotypes in animal models has proven challenging, necessitating the development of alternatives. We generated human induced pluripotent stem cell (iPSC)-derived retinal organoids of patients with retinitis pigmentosa caused by biallelic CRB1 mutations and evaluated them against autologous gene-corrected hiPSCs and hiPSCs from healthy individuals. Patient organoids show decreased levels of CRB1 and NOTCH1 expression at the retinal outer limiting membrane. Proximity ligation assays show that human CRB1 and NOTCH1 can interact via their extracellular domains. CRB1 patient organoids feature increased levels of WDFY1+ vesicles, fewer RAB11A+ recycling endosomes, decreased VPS35 retromer complex components, and more degradative endolysosomal compartments relative to isogenic control organoids. Taken together, our data demonstrate that patient-derived retinal organoids enable modeling of retinal degeneration and highlight the importance of CRB1 in early endosome maturation receptor recycling in the retina.

AAV-mediated gene augmentation therapy of CRB1 patient-derived retinal organoids restores the histological and transcriptional retinal phenotype.

Retinitis pigmentosa and Leber congenital amaurosis are inherited retinal dystrophies that can be caused by mutations in the Crumbs homolog 1 (CRB1) gene. CRB1 is required for organizing apical-basal polarity and adhesion between photoreceptors and Müller glial cells. CRB1 patient-derived induced pluripotent stem cells were differentiated into CRB1 retinal organoids that showed diminished expression of variant CRB1 protein observed by immunohistochemical analysis. Single-cell RNA sequencing revealed impact on, among others, the endosomal pathway and cell adhesion and migration in CRB1 patient-derived retinal organoids compared with isogenic controls. Adeno-associated viral (AAV) vector-mediated hCRB2 or hCRB1 gene augmentation in Müller glial and photoreceptor cells partially restored the histological phenotype and transcriptomic profile of CRB1 patient-derived retinal organoids. Altogether, we show proof-of-concept that AAV.hCRB1 or AAV.hCRB2 treatment improved the phenotype of CRB1 patient-derived retinal organoids, providing essential information for future gene therapy approaches for patients with mutations in the CRB1 gene.

Sleep Deprivation Does not Change the Flash Electroretinogram in Wild-type and Opn4-/-Gnat1-/- Mice.

Sleep deprivation reduces the response of neuronal activity in the suprachiasmatic nucleus (SCN) and the phase shift in circadian behaviour to phase shifting light pulses, and thus seems to impair the adaptation of the circadian clock to the external light-dark cycle. The question remains where in the pathway of light input to the SCN the response is reduced. We therefore investigated whether the electroretinogram (ERG) changes after sleep deprivation in wild-type mice and in Opn4-/-Gnat1-/- mutant male mice. We found that the ERG is clearly affected by the Opn4-/-Gnat1-/- mutations, but that the ERG after sleep deprivation does not differ from the baseline response. The difference between wild-type and mutant is in accordance with the lack of functional rod and melanopsin in the retina of the mutant mice. We conclude that the decrease in light responsiveness of the SCN after sleep deprivation is probably not caused by changes at the retinal level, but rather at the postsynaptic site within the SCN, reflecting affected neurotransmitter signalling.

Defining Phenotype, Tropism, and Retinal Gene Therapy Using Adeno-Associated Viral Vectors (AAVs) in New-Born Brown Norway Rats with a Spontaneous Mutation in Crb1.

Mutations in the Crumbs homologue 1 (CRB1) gene cause inherited retinal dystrophies, such as early-onset retinitis pigmentosa and Leber congenital amaurosis. A Brown Norway rat strain was reported with a spontaneous insertion-deletion (indel) mutation in exon 6 of Crb1. It has been reported that these Crb1 mutant rats show vascular abnormalities associated with retinal telangiectasia and possess an early-onset retinal degenerative phenotype with outer limiting membrane breaks and focal loss of retinal lamination at 2 months of age. Here, we further characterized the morphological phenotype of new-born and adult Crb1 mutant rats in comparison with age-matched Brown Norway rats without a mutation in Crb1. A significantly decreased retinal function and visual acuity was observed in Crb1 mutant rats at 1 and 3 months of age, respectively. Moreover, in control rats, the subcellular localization of canonical CRB1 was observed at the subapical region in Müller glial cells while CRB2 was observed at the subapical region in both photoreceptors and Müller glial cells by immuno-electron microscopy. CRB1 localization was lost in the Crb1 mutant rats, whereas CRB2 was still observed. In addition, we determined the tropism of subretinal or intravitreally administered AAV5-, AAV9- or AAV6-variant ShH10Y445F vectors in new-born control and Crb1 mutant rat retinas. We showed that subretinal injection of AAV5 and AAV9 at postnatal days 5 (P5) or 8 (P8) predominantly infected the retinal pigment epithelium (RPE) and photoreceptor cells; while intravitreal injection of ShH10Y445F at P5 or P8 resulted in efficient infection of mainly Müller glial cells. Using knowledge of the subcellular localization of CRB1 and the ability of ShH10Y445F to infect Müller glial cells, canonical hCRB1 and hCRB2 AAV-mediated gene therapy were explored in new-born Crb1 mutant rats. Enhanced retinal function after gene therapy delivery in the Crb1 rat was not observed. No timely rescue of the retinal phenotype was observed using retinal function and visual acuity, suggesting the need for earlier onset of expression of recombinant hCRB proteins in Müller glial cells to rescue the severe retinal phenotype in Crb1 mutant rats.

AAV-CRB2 protects against vision loss in an inducible CRB1 retinitis pigmentosa mouse model.

Loss of Crumbs homolog 1 (CRB1) or CRB2 proteins in Müller cells or photoreceptors in the mouse retina results in a CRB dose-dependent retinal phenotype. In this study, we present a novel Müller cell-specific Crb1 KO Crb2 LowMGC retinitis pigmentosa mouse model (complete loss of CRB1 and reduced levels of CRB2 specifically in Müller cells). The Crb double mutant mice showed deficits in electroretinography, optokinetic head tracking, and retinal morphology. Exposure of retinas to low levels of dl-α-aminoadipate acid induced gliosis and retinal disorganization in Crb1 KO Crb2 LowMGC retinas but not in wild-type or Crb1-deficient retinas. Crb1 KO Crb2 LowMGC mice showed a substantial decrease in inner/outer photoreceptor segment length and optokinetic head-tracking response. Intravitreal application of rAAV vectors expressing human CRB2 (hCRB2) in Müller cells of Crb1 KO Crb2 LowMGC mice subsequently exposed to low levels of dl-α-aminoadipate acid prevented loss of vision, whereas recombinant adeno-associated viral (rAAV) vectors expressing human CRB1 (hCRB1) did not. Both rAAV vectors partially protected the morphology of the retina. The results suggest that hCRB expression in Müller cells is vital for control of retinal cell adhesion at the outer limiting membrane, and that the rAAV-cytomegalovirus (CMV)-hCRB2 vector is more potent than rAAV-minimal CMV (CMVmin)-hCRB1 in protection against loss of vision.

Recombinant Adeno-Associated Viral Vectors (rAAV)-Vector Elements in Ocular Gene Therapy Clinical Trials and Transgene Expression and Bioactivity Assays.

Inherited retinal dystrophies and optic neuropathies cause chronic disabling loss of visual function. The development of recombinant adeno-associated viral vectors (rAAV) gene therapies in all disease fields have been promising, but the translation to the clinic has been slow. The safety and efficacy profiles of rAAV are linked to the dose of applied vectors. DNA changes in the rAAV gene cassette affect potency, the expression pattern (cell-specificity), and the production yield. Here, we present a library of rAAV vectors and elements that provide a workflow to design novel vectors. We first performed a meta-analysis on recombinant rAAV elements in clinical trials (2007-2020) for ocular gene therapies. We analyzed 33 unique rAAV gene cassettes used in 57 ocular clinical trials. The rAAV gene therapy vectors used six unique capsid variants, 16 different promoters, and six unique polyadenylation sequences. Further, we compiled a list of promoters, enhancers, and other sequences used in current rAAV gene cassettes in preclinical studies. Then, we give an update on pro-viral plasmid backbones used to produce the gene therapy vectors, inverted terminal repeats, production yield, and rAAV safety considerations. Finally, we assess rAAV transgene and bioactivity assays applied to cells or organoids in vitro, explants ex vivo, and clinical studies.

Human iPSC-Derived Retinas Recapitulate the Fetal CRB1 CRB2 Complex Formation and Demonstrate that Photoreceptors and Müller Glia Are Targets of AAV5.

Human retinal organoids from induced pluripotent stem cells (hiPSCs) can be used to confirm the localization of proteins in retinal cell types and to test transduction and expression patterns of gene therapy vectors. Here, we compared the onset of CRB protein expression in human fetal retina with human iPSC-derived retinal organoids. We show that CRB2 protein precedes the expression of CRB1 in the developing human retina. Our data suggest the presence of CRB1 and CRB2 in human photoreceptors and Müller glial cells. Thus the fetal CRB complex formation is replicated in hiPSC-derived retina. CRB1 patient iPSC retinal organoids showed disruptions at the outer limiting membrane as found in Crb1 mutant mice. Furthermore, AAV serotype 5 (AAV5) is potent in infecting human Müller glial cells and photoreceptors in hiPSC-derived retinas and retinal explants. Our data suggest that human photoreceptors can be efficiently transduced by AAVs in the presence of photoreceptor segments.

Production of iPS-Derived Human Retinal Organoids for Use in Transgene Expression Assays.

In vitro retinal organoid modeling from human pluripotent stem cells is becoming more common place in many ophthalmic laboratories worldwide. These organoids mimic human retinogenesis through formation of organized layered retinal structures that display markers for typical retinal cell types. Pivotally these humanized retinal models provide a stepping stone to the clinic as therapeutic tools and are expected to provide a promising alternative to current animal models. Thus pluripotent stem cell based healthy as well as diseased human retinal organoids are attractive for use in drug potency assays and gene augmentation therapeutics. Here we outline an established protocol for generation of these retinal organoids and how they can be used in conjunction with adeno-associated virus vectors for transgene expression assays.

AAV Serotype Testing on Cultured Human Donor Retinal Explants.

This protocol details on a screening method for infectivity and tropism of different serotypes of adeno-associated viruses (AAVs) on human retinal explants with cell-type specific or ubiquitous green fluorescent protein (GFP) expression vectors. Eyes from deceased adult human donors are enucleated and the retinas are isolated. Each retina is punched into eight to ten 6-mm equal pieces. Whatman™ paper punches are placed on the retinas and the stack is transferred onto 24-well culture inserts with the photoreceptors facing the membrane. AAVs are applied on the retinal explant punches to allow transduction for 48 h. Retinas are nourished by a serum-free Neurobasal®-A based medium composition that allows extended culturing of explants containing photoreceptor inner and outer segments. The protocols include quality control measurements and histological staining for retina cells. The cost and time effective procedure permits AAV transgene expression assays, RNAi knockdown, and pharmacological intervention on human retinas for 21 days ex vivo.