Abstract
Purpose
Biallelic pathogenic variants in the RCC1 and BTB domain-containing protein 1 (RCBTB1) gene cause an adult-onset retinal dystrophy. Here, we generated a knockout mouse model of RCBTB1 deficiency for the evaluation of RCBTB1 gene therapy.
Methods
Rcbtb1-knockout (KO) mice were generated with a homozygous deletion removing exons 2 and 3 of Rcbtb1. Wild-type (WT) and Rcbtb1-KO mice were assessed by optical coherence tomography and electroretinography at 3, 8, and 14 months of age. Retinal ultrastructure was assessed by transmission electron microscopy. Subretinal injections of adeno-associated virus 2 (AAV2)–RCBTB1 or AAV2–enhanced green fluorescent protein (EGFP) vector were performed at 2 months, and mice were analyzed at 8 months. Retinal gene expression was assessed by quantitative PCR and immunohistochemistry.
Results
Retinal Rcbtb1 expression was absent in Rcbtb1-KO mice. Eight-month-old Rcbtb1-KO mice showed reduced outer retinal thickness compared with WT mice. Ultrastructural analysis demonstrated increased mitochondrial damage in retinal pigment epithelial cells and increased frequencies of mitochondria with oxidative inclusions in photoreceptor inner segments in 8-month-old Rcbtb1-KO mice. Degenerating retinal pigment epithelium (RPE) and photoreceptors were observed in Rcbtb1-KO mice. Bruch's membrane appeared thicker in Rcbtb1-KO mice and contained druse-like deposits. Treatment with AAV2–RCBTB1 induced sustained RCBTB1 expression and preserved outer retinal thickness in 8-month-old Rcbtb1-KO mice.
Conclusions
Rcbtb1-KO mice showed accelerated outer retinal thinning, increased mitochondrial damage in the RPE, and photoreceptor apoptosis. AAV2–RCBTB1 vectors induced long-term retinal expression of RCBTB1 and prevented retinal thinning in Rcbtb1-KO mice.
Translational Relevance
Rcbtb1-KO mice provide a useful animal model for modeling RCBTB1 deficiency and preclinical screening of novel treatments.
Keywords: Rcbtb1, inherited retinal disease, knockout mouse, oxidative stress, gene therapy
Introduction
RCBTB1-associated retinopathy is an autosomal recessive inherited retinal disease caused by inheritance of biallelic pathogenic variants in the RCC1 and BTB domain-containing protein 1 (RCBTB1) gene. Currently, 22 patients with biallelic RCBTB1 mutations have been reported with retinal phenotypes ranging from a rod–cone dystrophy to a progressive macular chorioretinal atrophy beginning in third to fifth decade of life1–5 Our multimodal imaging natural history study of a patient with RCBTB1-associated macular chorioretinal atrophy demonstrated central regions of retinal pigment epithelium (RPE) atrophy that expanded centrifugally, followed by secondary loss of photoreceptors,3 a clinical phenotype previously associated with type 3 mitochondrial retinopathies.6 Previously, it was suggested that dominant RCBTB1 mutations were associated with familial exudative vitreoretinopathy and Coats disease7; however, a subsequent study on a larger cohort of 29 families found no association between the inheritance of single pathogenic RCBTB1 variants and retinal disease.4,8
Investigations by several groups have led to important insights into the molecular consequences of RCBTB1 deficiency. The RCBTB1 gene is comprised of 12 exons located on chromosome 13q14.2, and it encodes the 58-kDa RCBTB1 mRNA that is ubiquitously expressed in all tissue types, including the retina, where protein expression has been detected primarily in the inner retinal layers in healthy human and mouse eyes.2 The RCBTB1 protein incorporates six N-terminal regulator of chromosome condensation 1 (RCC1)-like domains (RLDs) and the C-terminal broad-complex, tramtrack, and bric-à-brac (BTB) domains.9 It has been shown to interact with components of the cullin-3 ubiquitin ligase complex, suggesting a role as a substrate adaptor protein in the ubiquitination pathway.10 Reduced expression of nuclear factor erythroid 2-related factor 2 (NRF2) antioxidant pathway genes was observed in lymphocytes derived from patients with RCBTB1-associated retinopathy2 and in a Rcbtb1 knockout (KO) frog model (Carron M, et al. IOVS. 2020;61:ARVO Abstract 1125). Our group previously demonstrated that RPE generated from induced pluripotent stem cells (iPSCs) derived from an RCBTB1-associated retinopathy patient11 showed increased levels of reactive oxygen species (ROS) and reduced activation of the NRF2 pathway.12 Patient-derived RPE cells also showed decreased primary cilia lengths and reduced mitochondrial function, as well as increased sensitivity to an oxidative stress challenge.13 Additionally, RCBTB1 mRNA expression was upregulated in healthy iPSC-derived RPE following an oxidative stress challenge. Together, these studies highlight the role of RCBTB1 in NRF2-pathway activation and implicate oxidative stress as a primary driver of disease pathogenesis in the RCBTB1-deficient retina.
To restore RCBTB1 expression in RCBTB1-deficent retinal cells, we previously developed an adeno-associated virus 2 (AAV2)-based gene replacement therapy (AAV2–RCBTB1). Treatment of patient iPSC-derived RPE cells with AAV2–RCBTB1 partially restored the NRF2 response to oxidative stress, reducing ROS levels and increasing primary cilia lengths.12 In the present study, we characterized a KO mouse model of RCBTB1-associated retinopathy and investigated the use of this model for preclinical evaluation of the AAV2–RCBTB1 gene therapy vector.
Materials and Methods
Ethics
The use of animals and all procedures described in this study were approved by the Harry Perkins Institute of Medical Research Animal Ethics Committee (AE219) and adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.
Generation of Rcbtb1-KO Mice
Rcbtb1-KO mice were generated by Ozgene (Perth, WA, Australia) on a C57BL/6J background. Briefly, the Rcbtb1 genomic locus in mouse embryonic stem (mES) cells was modified to introduce loxP sites flanking exons 2 and 3, as well as a neomycin resistance gene cassette flanked by Flp recombinase target (FRT) sites in intron 3. Rcbtb1+/loxPneo mES cells were injected into goGermline blastocysts14 to produce Rcbtb1+/loxPneo mice. Rcbtb1+/loxPneo mice were then crossed with transgenic mice expressing Cre recombinase, producing Cre±/Rcbtb1+/ΔEx2-3 offspring in which exons 2 and 3 were removed by recombination of the flanking loxP sites. Cre±/Rcbtb1+/ΔEx2-3 mice were backcrossed with wild-type (WT) C57BL/6J mice to remove the Cre transgene, then self-crossed to generate homozygous Rcbtb1ΔEx2-3/ΔEx2-3 (Rcbtb1-KO) mice (Fig. 1). Homozygous Rcbtb1-KO mouse breeding colonies were then established to supply mice for this project. Genotyping was performed by quantitative PCR (qPCR) using primers listed in Supplementary Table S1 (Ozgene). Genotyping primers were designed to amplify the loxP scar inserted during generation of the knockout Rcbtb1 allele, which was not present in the WT allele. Controls omitting template DNA and mouse DNA with a known genotype were included in each qPCR run. The qPCR results were normalized to the mouse Tert housekeeping gene. Representative results are shown in Supplementary Figure S1. WT C57BL/6J mice were also provided by Ozgene. Mice were transported to and housed at the Harry Perkins Institute North Bioresources Facility.
Figure 1.
Generation of Rcbtb1-KO mice. (A) Maps showing (top to bottom) the mouse Rcbtb1 locus (exons are shown as boxes, coding sequence are indicated by peach shading; introns are not to scale); the loxPneo-targeting cassette (neomycin resistance genes are indicated in yellow); the Rcbtb1loxPneo locus after insertion of the loxPneo cassette in mouse embryonic stem cells; and the Rcbtb1ΔEx2-3 locus after excision of the loxPneo cassette by Cre recombinase. (B) Bar chart shows the expression levels of Rcbtb1 mRNA measured by qPCR of retinal cDNA samples from 8-month-old WT (n = 4) and Rcbtb1-KO (n = 6) eyes. Bars represent mean expression values normalized to WT controls. Error bars show standard deviations. Statistical significance was analyzed by LMM. *P < 0.05. (C, D) Merged fluorescence images showing retinal immunostaining of 14-month-old WT mice (C) and Rcbtb1-KO mice (D) using anti-Rcbtb1 antibodies (green signal). The blue signal shows nuclear counterstaining with DAPI. GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; PRL, photoreceptor layer; RPE, retinal pigment epithelium. Lower panels show enlarged views of the RPE and PRL, demonstrating autofluorescent material in the PRL with little immunoreactivity in the RPE layer. Scale bars: 50 µm.
Subretinal Injections
Generation of AAV2–RCBTB1 and AAV2–enhanced green fluorescent protein (EGFP) vectors (Vector Biolabs, Malvern, PA) was previously described.12 For both AAV2/2 vectors, expression of transgenes with woodchuck hepatitis virus post-transcriptional response elements was driven by a CAG promoter. Subretinal injections were performed in 2-month-old mice as previously described.15 Briefly, mice were anesthetized with ketamine/xylazine (80/10 mg/kg) and placed on an operating microscope. Pupils were dilated by application of 1% tropicamide eye drops (Alcon, Fort Worth, TX). Then, 1 µL of AAV2–RCBTB1 vector or AAV2–EGFP vector (5 × 1012 genome copies/mL in phosphate-buffered saline [PBS] with 5% glycerol) was injected into the subretinal space of both eyes using a UMP3 UltraMicro Pump and a NanoFil 10-µL syringe (World Precision Instruments, Sarasota, FL) with a blunt 35-gauge needle. After injection, mice were imaged by optical coherence tomography (OCT) to confirm the subretinal bleb. Eyes showing severe surgical injury or unresolved retinal detachments by OCT at 6 months post-treatment were excluded from the analysis.
Optical Coherence Tomography
Retinal imaging was performed on anesthetized mice using the Bioptigen Envisu 2200 VHR Spectral Domain Ophthalmic Imaging System (SDOIS; Bioptigen, Morrisville, NC). Retinal thickness measurements were made on OCT B-scans taken 0.3 mm superior to the optic nerve head using ImageJ 1.53i (National Institutes of Health, Bethesda, MD). For each eye, mean retinal thickness was calculated by averaging measurements at three locations (nasal, central, and temporal) across the OCT fundus B-scans.
Electroretinography
Electroretinography (ERG) was performed under scotopic and photopic conditions on anesthetized mice using a Celeris full-field ERG system (Diagnosys, Lowell, MA), as previously described.16 For scotopic ERGs, mice were dark adapted overnight, and measurements were performed under dim red lighting. Mice were subjected to 1-ms flashes with intensities of 0.01, 0.1, 0.3, 1, 3, 10, and 25 cd·s/m2. Each flash was repeated four times at 0.10 Hz at 10-second intervals, with 60-second intervals between different flash intensities. Following scotopic ERG readings, mice were light adapted for 10 minutes at 30 cd·s/m2. Mice were then presented with a series of flashes on a background of 30 cd/m2 at 2 Hz and at intensities of 3 and 10 cd·s/m2. ERG trace data were averaged across all eyes within each group. ERG a-wave and b-wave amplitudes were calculated using Espion V6 software (Diagnosys).
RNA Analysis
Following euthanasia, mouse eyes were enucleated and retinas were dissected into TRIzol (Thermo Fisher Scientific, Waltham, MA) for total RNA extraction according to the manufacturer's protocol. Retinal RNA was quantified using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific), then converted to cDNA using the RT2 First Strand Kit (QIAGEN, Hilden, Germany). Quantitative PCR was performed on a Bio-Rad CFX Connect Real-Time PCR System (40 cycles, 95°C for 20 seconds, 60°C for 60 seconds; Bio-Rad Laboratories, Hercules, CA). Gene expression values were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression using the ΔCT method. Primers used in this study are listed in Supplementary Table S1.
Immunohistochemistry
Following euthanasia, mouse eyes were marked with an orientation reference at the superior cornea using a needle puncture, then enucleated and transferred to PBS containing 4% paraformaldehyde for 2 days at 4°C. Eyes were then transferred to PBS with 30% sucrose for a further 2 days at 4°C. After fixation and cryoprotection, cells were embedded in optimal cutting temperature compound (Sakura Finetek USA, Torrance, CA) and frozen at –80°C. Then, 8-µm cryosections were prepared on a Leica CM3050 S cryostat (Leica Microsystems, Wetzlar, Germany). Cryosections were incubated with blocking buffer (PBS, 5% goat serum, and 0.3% Triton X-100) for 1 hour at room temperature. Primary antibodies, including rabbit RCBTB1 Polyclonal Antibody (PA5-30672, RRID AB_2548146, 1:100; Thermo Fisher Scientific), mouse anti-8-hydroxyguanosine (8OHG; DNA/RNA Oxidative Damage Mouse Monoclonal Antibody, 12501, RRID AB_129849, 1:100; Sapphire Bioscience, Redfern, NSW, Australia), and rabbit GFP Polyclonal Antibody (A-11122, RRID AB_221569, 1:200; Thermo Fisher Scientific), were diluted in blocking buffer and added to the cells for overnight incubation at 4°C. Cells were then washed three times with PBS before the addition of fluorophore-conjugated secondary antibodies, including Goat anti-Rabbit IgG (H+L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor 488 (A-11034, RRID AB_2576217, 1:500; Thermo Fisher Scientific), Goat anti-Mouse IgG (H+L) Highly Cross-Adsorbed Antibody, Alexa Fluor 546 (A-11030, RRID AB_144695, 1:500; Thermo Fisher Scientific), and the nuclear stain 4′,6-diamidino-2-phenylindole (DAPI; 1 µg/mL; Sigma-Aldrich, St. Louis, MO). After 1 hour, slides were then washed in PBS and coverslips mounted with Fluoromount Aqueous Mounting Media (Sigma-Aldrich). Slides were imaged at 200× to 400× magnification on a Nikon ECLIPSE Ni series fluorescence microscope (Nikon Corporation, Tokyo, Japan).
Analysis of DNA/RNA Oxidative Damage
One section from three WT and three Rcbtb1-KO mice was immunostained for 8OHG, as described above. A negative control omitting primary antibodies was used to calibrate image acquisition settings. Retinal immunolabeling intensities were measured using ImageJ 1.44o.
Transmission Electron Microscopy
Single eyes from two WT and two KO mice were enucleated, punctured in the superior cornea, and fixed in in 2.5% (v/v) phosphate buffered glutaraldehyde for a minimum of 48 hours. Using the Lynx II Automated Tissue Processor (Electron Microscopy Sciences, Hatfield, PA), fixed eyes were rinsed in saline and then immersed in 1% aqueous osmium tetroxide before dehydration in a graded series of ethanol concentrations. The tissue was then infiltrated with propylene oxide, followed by increasing concentrations of propylene oxide and araldite resin, then infiltrated with three changes of pure araldite resin. Blocks were cured for 24 hours at 80°C. Mouse eye blocks were sectioned with an RMC Ultramicrotome (Boeckeler Instruments, Tucson, AZ). Blocks were trimmed until the optic nerve was visible, then 1 µm sections were cut, mounted on glass slides and stained with 0.1% methylene blue (ProSciTech, Kirwan, QLD, Australia). Methylene blue sections were imaged at 200× to 400× magnification on a Nikon ECLIPSE Ni series fluorescence microscope. For transmission electron microscopy (TEM), two sections approximately 95 nm thick were then cut with an RMC Ultramicrotome and mounted on copper grids (ProSciTech) before staining with uranyl acetate (ProSciTech) and lead citrate solutions (ProSciTech) with distilled water rinses in between. Grids were examined and photographed with a JEM-1400 transmission electron microscope (JEOL, Tokyo, Japan) fitted with an ORIUS 11-Mp digital camera (Orius SC1000; Gatan, Pleasanton, CA). Digital images were converted to TIFF files and analyzed using ImageJ. For mitochondrial analyses, images of RPE and photoreceptor inner segments were collected at 30,000× magnification from retinal regions approximately 1 mm from the optic nerve head. Mitochondrial length (longest axis) and width (shortest axis) were measured in ImageJ. The mitochondrial aspect ratio was calculated as mitochondrial length/width. Cristae grading in RPE and inclusion counting in the inner segment were performed as described in the Results section.
Statistical Analysis
For OCT and ERG data, which included both eyes from each mouse, statistical significance was assessed using linear mixed modeling (LMM) in R 4.4.3 (GUI 1.81; R Foundation for Statistical Computing, Vienna, Austria). Models were fitted with the lme4 package, with genotype, age, and eye specified as fixed effects (including their interactions) and mouse included as a random intercept to account for repeated measures. Significance of fixed effects was assessed using the lmerTest package, with degrees of freedom estimated by the Kenward–Roger method. Post hoc pairwise comparisons were performed with the emmeans package, using Tukey adjustment for multiple testing with statistical significance defined as P < 0.05. Experiments including a single eye from each mouse were analyzed using unpaired t-tests in Prism 9.5.1 (GraphPad Software, Boston, MA), with P < 0.05 considered significant.
Results
Generation of Rcbtb1-KO Mice
Rcbtb1-KO mice were engineered on a C57BL/6J background to carry homozygous Rcbtb1 alleles in which exons 2 and 3 had been deleted, removing the start codon from the Rcbtb1 transcript (Fig. 1A). Homozygous Rcbtb1-KO mice were viable and fertile. Litter sizes and overall health of homozygous Rcbtb1-KO breeding colonies were comparable with the C57BL/6J WT control strain. Analysis of retinal Rcbtb1 mRNA expression by qPCR demonstrated a significant 99.92% reduction in Rcbtb1-KO mice, compared with age-matched WT mice (Fig. 1B). In WT mice, retinal Rcbtb1 immunoreactivity was predominantly localized to the inner retina (including the ganglion cell, inner plexiform, and outer plexiform layers) but was not detected in the RPE cell layer. Photoreceptor outer segments showed some autofluorescence in both WT and KO mice at 14 months of age. Rcbtb1-KO mice showed an absence of retinal Rcbtb1 immunoreactivity (Fig. 1C), demonstrating the successful knockout of Rcbtb1 gene expression in the mouse.
Accelerated Outer Retinal Thinning in Rcbtb1-KO Mice
Retinal architecture appeared healthy in OCT images from both WT and Rcbtb1-KO mice (Fig. 2A). Retinal thickness measurements were taken from fundus OCT images in WT and Rcbtb1-KO mice at 3, 8, and 14 months of age (Fig. 2B). Total neuroretinal thickness was not significantly different between genotypes at any age. No significant changes in total neuroretinal thickness were detected in WT or Rcbtb1-KO mice between 3 and 8 months of age; however, both strains showed significant age-dependent thinning of the neuroretina between 8 and 14 months of age (P < 0.001) (Fig. 2C). No significant changes in RPE layer thicknesses were detected with age or between genotypes (Fig. 2D).
Figure 2.
Retinal thickness and function in Rcbtb1-KO mice. (A) Representative en face and B-scan OCT images showing the retinal fundus of 8-month-old WT and Rcbtb1-KO mice. The green line indicates the location of B-scans used for retinal thickness measurements. (B) OCT B-scan images from 8-month-old WT and Rcbtb1-KO mice. Borders for measurement of total (NR, yellow line), inner (IR, blue line), and outer (OR, green line) neuroretina and RPE (red line) layers are indicated. RGC, retinal ganglion cell; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; ELM, external limiting membrane; IS/OS, inner/outer segments; RPE, retinal pigment epithelium; BM, Bruch's membrane. (C–F) Bar charts showing mean thicknesses of NR (C), RPE (D), OR (E), and IR (F) in WT and Rcbtb1-KO mice at 3 months of age (n = 12 WT and 12 KO eyes), 8 months of age (n = 11 WT and 10 KO eyes), and 14 months of age (n = 8 WT and 10 KO eyes). Error bars indicate SEM. Data were analyzed by LMM. *P < 0.05. (G–I) Scotopic ERGs were performed in dark-adapted WT and Rcbtb1-KO mice at 8 months of age (n = 10 WT and 12 KO eyes) and 14 months of age (n = 8 WT and 10 KO eyes). Graphs show group-averaged scotopic ERG traces (G), and bar charts depict mean scotopic a-wave (H) and b-wave (I) amplitudes following a 0.01 cd·s/m2 flash. Error bars indicate SEM. (J–L) Photopic ERGs were performed in the same mice after light adaptation. Graphs show group-averaged photopic ERG traces (J), and bar charts depict mean photopic a-wave (K) and b-wave (L) amplitudes following a 10 cd·s/m2 flash. Error bars indicate SEM. Data were analyzed by LMM. *P < 0.05, **P < 0.01, ***P < 0.001; ns, not significant (P > 0.05).
Significant age-related reductions in outer retinal thicknesses (including the outer nuclear and inner and outer photoreceptor segment layers) were detected in both WT (P < 0.0001) and Rcbtb1-KO (P = 0.0024) mice between 8 and 14 months of age. At 8 months, Rcbtb1-KO mice showed a significant reduction in outer retinal thickness compared with WT controls (P = 0.0275) (Fig. 2E). Significant reductions in inner retinal thicknesses (including the ganglion cell, inner plexiform, inner nuclear, and outer plexiform layers) were also observed in both WT and Rcbtb1-KO mice with increasing age; however, no significant differences were detected between genotypes (Fig. 2F). Together, these results suggest that age-related outer retinal thinning is accelerated in Rcbtb1-KO mice.
Late-Onset Rod Dysfunction in Rcbtb1-KO Mice
Scotopic and photopic ERGs were performed in WT and Rcbtb1-KO mice at 8 and 14 months of age. Scotopic a-wave amplitudes following a 0.01 cd·s/m2 flash were similar between WT and Rcbtb1-KO mice at 8 months of age (Figs. 2G–I). However, by 14 months of age, Rcbtb1-KO mice showed significantly reduced a-wave amplitudes compared with age-matched WT control mice (P = 0.0465). Additionally, 14-month-old Rcbtb1-KO mice showed significant reductions in a-wave amplitude compared with 8-month-old Rcbtb1-KO mice (P = 0.005), but WT mice showed only a small, non-significant reduction in a-wave amplitude between 8 and 14 months of age (P = 0.2230). Scotopic b-wave amplitudes showed no significant differences between genotypes; however, Rcbtb1-KO mice showed a significant reduction in b-wave amplitude between 8 and 14 months of age (P = 0.0016), but WT mice showed no significant reduction in b-wave amplitude with increasing age (P = 0.2005) (Figs. 2G–I).
Photopic ERG traces were similar in WT and Rcbtb1-KO mice following a 10 cd·s/m2 flash up to 14 months of age (Figs. 2J–L). Neither age nor genotype had any statistically significant effect on photopic a-wave or b-wave amplitudes (P > 0.05). Similar results were observed using a 3 cd·s/m2 flash (data not shown). Together, these results demonstrate accelerated age-related decline in rod-driven scoptopic function in Rcbtb1-KO mice, whereas cone-driven photopic retinal function remained relatively stable.
Increased Oxidative Stress in the Aged Rcbtb1-KO Retina
To investigate oxidative stress in the Rcbtb1-KO retina, we performed immunostaining for 8OHG, a marker of DNA/RNA oxidative damage. The 14-month-old WT mice showed low levels of 8OHG immunoreactivity in the retinal ganglion cell layer and outer plexiform layer (Fig. 3A). In contrast, 14-month-old Rcbtb1-KO mice showed increased 8OHG immunoreactivity, with prominent labeling observed across the retina (Fig. 3B). Analysis of immunostaining intensities showed that retinal 8OHG immunoreactivity was significantly increased in 14-month-old Rcbtb1-KO mice compared with age-matched WT mice (P = 0.025) (Fig. 3C).
Figure 3.
Retinal DNA/RNA oxidative damage in aged Rcbtb1-KO Mice. (A, B) Representative images following immunostaining for 8OHG (red signal) on retinal sections from 14-month-old WT mice (A) and Rcbtb1-KO mice (B) showing increased 8OHG immunoreactivity in Rcbtb1-KO mice. Blue signal shows nuclear counterstaining with DAPI. Scale bars: 50 µm. (C) Quantification of 8OHG immunoreactivity showed a significant twofold increase in DNA/RNA oxidative damage across the retina in Rcbtb1-KO mice. Mean immunostaining intensities were calculated from three WT and three KO mice. Error bars indicate standard deviation. Statistical comparison was performed by t-test. *P < 0.05.
Rcbtb1-KO Mice Display Ultrastructural Pathology of the RPE and Bruch's Membrane
TEM analysis demonstrated similar RPE ultrastructure in 8-month-old WT and Rcbtb1-KO mice, with comparable pigment granules, apical cilia densities, and basal infoldings (Figs. 4A–D). However, Bruch's membrane appeared thickened in the Rcbtb1-KO mouse compared to the WT (Figs. 4E, 4F), and large drusen-like deposits were identified beneath RPE cells (Fig. 4G) and within Bruch's membrane (Fig. 4H) in Rcbtb1-KO TEM sections. RPE mitochondrial lengths and widths were measured and cristae morphology graded according to the scale demonstrated in Figure 5A. Both WT and KO RPE cells showed abundant basal mitochondria with cristae morphologies ranging from healthy, with dense cristae formations throughout the majority of the mitochondrial interior, to damaged, with few or no visible cristae. However, although 88% of WT RPE mitochondria showed healthy cristae morphology, only 52% of Rcbtb1-KO RPE mitochondria were healthy, and 48% showed damaged ultrastructure (Fig. 5A). The frequency of damaged mitochondria (with a cristae grade of 0–1) was significantly increased in 8-month-old KO mice compared with 8-month-old WT (P < 0.05) (Fig. 5B). Mean cristae grade was significantly decreased in KO RPE mitochondria (P < 0.001) (Fig. 5C). RPE mitochondrial lengths were similar in 8-month-old WT and KO mice (Fig. 5D); however, mitochondrial widths were significantly decreased (Fig. 5E), and aspect ratios were significantly increased in KO mice (Fig. 5F), suggesting potential elongation of Rcbtb1-deficient RPE mitochondria. These results demonstrate increased mitochondrial damage in the RPE layer of Rcbtb1-KO mice.
Figure 4.
Ultrastructural analysis of retinal pigment epithelium. (A–F) TEM images showing RPE (A–D) and Bruch's membrane (E–F) ultrastructure in 8-month-old WT mice (A, C, E) and Rcbtb1-KO mice (B, D, F). OS, outer segment; MV, apical microvilli; BI, basal infoldings; M, mitochondria; BM, Bruch's membrane; BMD, Bruch's membrane deposit; N, nucleus; DD, drusen-like deposit. Scale bars: 10 µm (A, B); 5 µm (C, D); 1 µm (E, F). (G) A large drusen-like deposit identified in an Rcbtb1-KO retinal section. Scale bar: 5 µm. The red box indicates the region shown in panel (H). (H) A drusen-like deposit located within Bruch's membrane was identified in an Rcbtb1-KO retinal section, comprised of granular material interspersed with degenerate organelle structures. Scale bar: 500 nm.
Figure 5.
Analysis of RPE mitochondria. (A) Representative TEM images show RPE mitochondria from 8-month-old WT (upper panels; n = 99) and Rcbtb1-KO mice (lower panels; n = 108) with different grades of cristae morphology, ranging from healthy (3–2) to damaged (1–0). Insets indicate the percentage of mitochondria in each grade. Rcbtb1-KO mice showed higher proportions of damaged RPE mitochondria than WT mice. (B) Bar graph shows mean percentages of damaged mitochondria in WT and Rcbtb1-KO RPE. Error bars indicate standard deviation. Significance was assessed by t-test. *P < 0.05. Rcbtb1-KO mice showed significantly higher proportions of damaged RPE mitochondria than WT mice. (C–F) Mean mitochondrial cristae grade (E), length (D), width (E), and aspect ratio (length/width) (F) were calculated across all RPE mitochondria measured in 8-month-old WT and Rcbtb1-KO mice. Error bars indicate standard deviation. Rcbtb1-KO mice showed significantly reduced cristae grade, width, and aspect ratios compared with WT mice. Data were analyzed by t-test. ***P < 0.001, ****P < 0.0001; ns, not significant (P > 0.05).
Rcbtb1-KO Mice Display Ultrastructural Pathology in Photoreceptor Cells
We further examined the ONL of 8-month-old WT and Rcbtb1-KO mice by methylene blue staining and TEM histology. Consistent with the OCT results, histological analysis of methylene blue–stained retinal sections from 8-month-old WT and Rcbtb1-KO mice demonstrated reduced ONL thickness and nuclear densities in KO mice compared with WT. The choroid and INL also appeared thinner in Rcbtb1-KO retinal sections, but the RPE had a similar appearance between genotypes (Supplementary Fig. S2).
TEM imaging of the WT ONL showed dense, regular columns of rod photoreceptor nuclei with highly condensed chromatin, interspersed by Müller glial processes and with frequent cone cell nuclei observed in the outermost layers of the ONL. In contrast, the ONL in Rcbtb1-KO mice appeared thinner and more disordered, with lower nuclear densities interspersed with electron-lucent vacuoles (Figs. 6A, 6B). Cone nuclei were less frequently observed in the ONL of KO mice (Figs. 6C, 6D); however, apoptotic cone nuclei were identified in the outermost layers of the ONL (Fig. 6E). Apoptotic rod nuclei were also identified throughout the ONL, often surrounded by vacuolated tissue (Figs. 6F–H). Quantification of cone (Fig. 6I) and pyknotic (Fig. 6J) nuclei in the ONL demonstrated a significant reduction in cone nuclei and significantly increased pyknotic nuclei in Rcbtb1-KO mice compared with WT mice.
Figure 6.
Ultrastructural analysis of the outer nuclear layer. (A, B) TEM images show the ONL of 8-month-old WT mice (A) and Rcbtb1-KO mice (B). The Rcbtb1-KO ONL appeared thinner and more disorganized, with vacuolated spaces. (C, D) Enlarged view of the outer layers of the ONL showing cone nuclei (CN) and rod nuclei (RN) in WT mice (C) and Rcbtb1-KO mice (D). Vacuoles were observed in Rcbtb1-KO mice (red arrows in D). (E, F) Examples of apoptotic cone nuclei (E) and rod nuclei (F) identified in the outer layers of the Rcbtb1-KO ONL (red arrows). (G, H) Examples of pyknotic rod nuclei associated with vacuolated spaces in the Rcbtb1-KO ONL. (I, J) Quantification of cone nuclei (I) and pyknotic nuclei (H) in retinal TEM images from two 8-month-old WT and two 8-month-old Rcbtb1-KO mice. Error bars indicate standard deviation. Statistical comparison was performed by t-test. *P < 0.05; **P < 0.01.
Mitochondria in the inner segments of photoreceptors of 8-month-old WT and KO mice showed elongated morphologies compared with RPE mitochondria. A proportion of inner segment mitochondria in both WT and KO mice contained inclusions, ranging from small electron-dense puncta to larger electron-lucent vesicles surrounded by electron-dense, membranous whorls (Figs. 7A–F). Rcbtb1-KO mice showed significantly increased frequencies of mitochondria containing inclusions (P > 0.05) (Fig. 7G). These inclusions have been previously associated with oxidative damage in photoreceptors,17 suggesting that Rcbtb1-KO–deficient inner segment mitochondria are under increased oxidative stress.
Figure 7.
Analysis of photoreceptor inner segment mitochondria. (A, B) TEM images show inner segments (ISs) and outer segments (OSs) in WT mice (A) and Rcbtb1-KO mice (B). (C–F) TEM images show mitochondria in the inner segments of 8-month-old WT mice (C, D) and Rcbtb1-KO mice (E, F). Mitochondria in both WT (n = 169) and Rcbtb1-KO (n = 139) ISs showed electron-dense inclusions and mitochondrial-derived vesicles (red arrows). (G) Bar graph shows mean percentages of IS mitochondria with inclusions in WT and Rcbtb1-KO mice. Error bars indicate standard deviation. Statistical significance was assessed by t-test. *P < 0.05. Rcbtb1-KO ISs showed significantly higher proportions of mitochondria with inclusions than in WT mice. (H–K) Mean mitochondrial cristae grade (H), length (I), width (J), and aspect ratio (length/width) (K) were calculated across all IS mitochondria measured in 8-month-old WT and Rcbtb1-KO mice. Error bars indicate standard deviation. Data were analyzed by t-test. *P < 0.05, **P < 0.01; ns, not significant (P > 0.05).
Inner segment mitochondria showed higher cristae grades than RPE mitochondria, and damaged mitochondria were rarely observed in both genotypes (Fig. 7H). Inner segment mitochondria showed significantly reduced lengths in KO mice compared with WT mice (Fig. 6I), but mitochondrial width was similar between genotypes (Figs. 7J, 7K). Given the length of inner segment mitochondria and the flexibility of this region during tissue processing, it is likely the differences in aspect ratio reflect subtle differences in obliquity between sections.
Sustained AAV2 Transgene Expression in the Mouse Retina
To evaluate RCBTB1 gene replacement therapy, we performed subretinal injections of AAV2–RCBTB1 vectors in WT and Rcbtb1-KO mice at 2 months. An additional group of WT mice received subretinal injections of AAV2–EGFP vectors. AAV2 vector constructs are shown in Figure 8A. Six months after injection (8 months of age), transgene expression was analyzed by qPCR and immunostaining and retinal thicknesses measured by OCT. Expression of human RCBTB1 or EGFP mRNA was sustained for up to 6 months after subretinal injection of AAV2 vectors (Fig. 8B; Supplementary Fig. S3). No significant difference in RCBTB1 expression was observed between WT and KO mice treated with AAV2–RCBTB1 (P = 0.8909). Immunohistochemistry using an anti-RCBTB1 antibody recognizing both mouse and human proteins demonstrated immunoreactivity in the retinae of untreated WT mice and WT and Rcbtb1-KO mice treated with AAV2–RCBTB1 6 months post-injection, but not in untreated age-matched Rcbtb1-KO mice (Fig. 8C). RCBTB1 immunoreactivity in Rcbtb1-KO mice treated with AAV2–RCBTB1 showed localization patterns in the inner retina similar to those seen in WT controls and was absent from the outer nuclear layer; however, mice treated with AAV2–RCBTB1 showed prominent RCBTB1 immunoreactivity in the photoreceptor outer segments and the RPE, which was not present in untreated Rcbtb1-KO or WT mice (Fig. 8C). Given the lack of RCBTB1 immunoreactivity in the outer nuclear layer and inner segments, it is possible that the apparent RCBTB1 transgene expression observed in the outer segments reflects increased autofluorescence in the outer segments of treated mice. Overall, these results demonstrate that AAV2 vectors achieved long-term retinal expression of EGFP and RCBTB1 in treated mice.
Figure 8.
AAV2–RCBTB1 gene replacement. (A) Schematic showing AAV2–RCBTB1 and AAV2–EGFP vectors, flanked by AAV inverted terminal repeats (ITRs). RCBTB1 or EGFP transgene expression was driven by a CAG promoter and included by a 3′ woodchuck hepatitis virus post-transcriptional regulatory element (WPRE) followed by a polyA tail (pA). (B) Bar graphs showing retinal expression levels of human RCBTB1 mRNA in 8-month-old untreated (UT) WT eyes, WT eyes treated with AAV2–RCBTB1 (AR) (n = 4 UT and 5 AR eyes), and Rcbtb1-KO mice (n = 3 UT and 6 AR eyes). Error bars indicate standard deviation. Data were analyzed by LMM. *P < 0.05; ***P < 0.001. Six months after subretinal injections, AAV2–RCBTB1 treatment significantly increased RCBTB1 expression in both WT and Rcbtb1-KO mice. (C) Merged fluorescence images showing anti-RCBTB1 (red signal) neuroretinal (upper panels) and RPE (lower panels) immunostaining of 8-month-old UT Rcbtb1-KO mice and AR-treated WT and Rcbtb1-KO mice. Blue signal shows nuclear counterstaining with DAPI. Scale bars: 50 µm. Labels indicate retinal layers. AAV2–RCBTB1 treatment restored RCBTB1 immunoreactivity in the inner retina of Rcbtb1-KO mice and induced RCBTB1 expression in the photoreceptor and RPE layers in WT and KO mice. (D) Bar charts showing mean outer retinal thicknesses in 8-month-old UT WT mice (n = 11), Rcbtb1-KO mice (n = 10), AAV2–EGFP (AE)-treated WT mice (n = 8), AR-treated WT mice (n = 11), and Rcbtb1-KO mice (n = 9). Error bars indicate SEM. *P < 0.05; **P < 0.01. Panels show representative OCT images for each group., with retinal layers indicated. AAV2–RCBTB1 treatment prevented ONL thinning in Rcbtb1-KO mice.
AAV2–RCBTB1 Treatment Preserves Outer Retinal Thickness in Adult Rcbtb1-KO Mice
Six months after injection, WT mice treated with AAV2–EGFP or AAV2–RCBTB1 showed no significant change in outer neuroretinal thicknesses compared with untreated WT mice (Fig. 8D). In contrast, outer neuroretinal thicknesses in Rcbtb1-KO mice treated with AAV2–RCBTB1 were significantly thicker (P < 0.0388) compared with untreated Rcbtb1-KO mice (Fig. 8D) and showed no significant change compared with untreated WT mice or mice treated with AAV2–EGFP or AAV2–RCBTB1. Together, these results demonstrate that AAV2–RCBTB1 treatments prevented outer retinal thinning in Rcbtb1-KO mice.
Discussion
In this study, we characterized the retinal phenotype of the Rcbtb1-KO mouse model of RCBTB1-associated retinopathy. Previous studies reporting the clinical phenotypes of RCBTB1-associated retinopathy patients demonstrated the development of regions of RPE atrophy, followed by secondary photoreceptor dysfunction and loss,1–4 highlighting similarities with geographic atrophy and retinal mitochondrial diseases.6,13 In our previous work, we identified macular volume and RPE atrophy as feasible endpoints for clinical monitoring of disease progression and showed reduced and delayed photopic a-waves and b-waves in an RCBTB1-associated retinopathy patient.3 Here, we examined the retinal phenotype of Rcbtb1-KO mice to identify correlates of disease for preclinical screening of potential therapeutics.
Retinal Phenotype of Rcbtb1-KO Mice
In their previous study, Coppieters et al.2 showed that Rcbtb1 immunoreactivity in the mouse retina was predominantly localized to the inner retinal layers but was not observed in photoreceptors or RPE cells. Here, we observed similar Rcbtb1 immunostaining patterns in WT mice, which showed strong inner retinal Rcbtb1 immunoreactivity that was not observed in the photoreceptor layers or the RPE. In contrast, Rcbtb1-KO mice showed no Rcbtb1 immunoreactivity in the retina, confirming successful knockout of the Rcbtb1 gene. These results suggest the Rcbtb1 protein is constitutively expressed in inner retinal cells in mice, predominantly localizing to the inner and outer plexiform layers and the ganglion cell layer.
Interestingly, Rcbtb1 immunoreactivity was not detected in the mouse RPE, which is thought to be the primary site of pathology in RCBTB1-associated retinopathy. In human iPSC-derived RPE cells, RCBTB1 expression was found to be low under resting conditions but was significantly upregulated following an oxidative stress challenge.13 Rcbtb1 mRNA expression has been previously demonstrated in mouse RPE cells.2 Together, these results suggest that Rcbtb1 protein expression is low in RPE cells under homeostatic conditions. Further studies are necessary to investigate whether expression of RCBTB1 protein is induced in the outer retina of mice in response to an oxidative stress challenge.
Because RCBTB1-associated retinopathy typically manifests in the third decade of life in human patients, we characterized Rcbtb1-KO mice and WT controls in early adulthood (3 months), middle adulthood (8 months), and old age (14 months). Significant thinning of all retinal layers was observed in mice of both genotypes between 8 months and 14 months, consistent with reported age-related retinal thinning in mice.18 At 8 months, outer retinal thickness was significantly reduced in Rcbtb1-KO mice compared with WT mice. Retinal thinning in Rcbtb1-KO mice was accompanied by ultrastructural changes in the ONL, with prominent vacuolated spaces, significantly increased frequencies of pyknotic nuclei, and reduced numbers of cone cell nuclei in the KO outer nuclear layer. However, the retinal thinning observed in the central retinal regions of Rcbtb1-KO mice was not associated with significant differences in fullfield ERG between WT and KO mice at 8 months of age, suggesting that photoreceptor losses across the whole retina are mild at this time point. Further studies are necessary to investigate rod and cone cell densities across different retinal regions.
By 14 months of age, Rcbtb1-KO mice showed a significant reduction in scotopic a-wave amplitudes compared with WT mice, consistent with a loss of rod photoreceptors. KO mice showed a significant reduction in scotopic a-wave and b-wave amplitudes between 8 and 14 months, but WT mice showed only a small, non-significant reduction, suggesting acceleration of age-related rod cell dysfunction in the Rcbtb1-deficient retina. Together, these results demonstrate that Rcbtb1-deficiency leads to middle-aged onset of photoreceptor loss in Rcbtb1-KO mice. Outer retinal thicknesses may provide a useful measure of disease onset and progression in mice from 8 months, and scotopic a-waves may provide a useful outcome measure in older mice.
Oxidative Stress in the Rcbtb1-KO Retina
Previous studies in patient-derived lymphocytes2 and patient iPSC-derived RPE12,13 showed increased intracellular levels of ROS and reduced activation of the NRF2 antioxidant pathway in RCBTB1-deficient cells. Reduced retinal NRF2 pathway activation was also reported in a Rcbtb1 knockout frog model (Carron M, et al., IOVS. 2020;61:ARVO Abstract 1125), suggesting a key role for RCBTB1 as a positive regulator in the NRF2 antioxidant defense pathway. We previously showed that RCBTB1 expression was increased in iPSC–RPE cells in response to an oxidative stress challenge and identified an antioxidant response element in the promoter of RCBTB1,13 providing a mechanism for positive feedback between NRF2 and RCBTB1 expression during activation of antioxidant defenses. These findings suggest that RCBTB1 is involved in activation of the NRF2 pathway and that impaired antioxidant defenses lead to increased oxidative stress in RCBTB1-deficient retinal cells. Here, we observed significantly increased 8OHG immunoreactivity in the retinae of aged Rcbtb1-KO mice compared with WT controls, consistent with increased levels of oxidative stress in the Rcbtb1-deficient mouse retina.
Additionally, we observed signs of increased retinal oxidative stress in 8-month-old Rcbtb1-KO mice by TEM. In photoreceptor inner segments, we frequently observed mitochondria containing electron-dense inclusions ranging from small dense puncta to larger electron-dense whorls surrounding electron-lucent vesicles. These inclusions resemble mitochondrial derived vesicles (MDVs) previously described in primate photoreceptors,17 which have been proposed to provide a mechanism for the removal of oxidized material from damaged mitochondria.20–22 Inclusions were present in approximately 15% of photoreceptor inner segment mitochondria in 8-month-old WT mice, and MDVs were also observed in the cytoplasm or subretinal space (data not shown). In contrast, 40% of inner segment mitochondria contained inclusions in 8-month-old Rcbtb1-KO mice. The increased MDV production observed in Rcbtb1-deficient photoreceptors is consistent with elevated mitochondrial oxidative stress levels in these cells.
MDVs were rarely observed in the RPE of either WT or Rcbtb1-KO mice; however, RPE mitochondria showed a wider range of cristae morphologies than photoreceptor mitochondria. Damaged mitochondria with few or no cristae accounted for approximately 12% of the RPE chondriome in WT mice, but almost half of the RPE mitochondria showed damaged cristae in the Rcbtb1-KO mice. These results mirror the increases in damaged mitochondria previously observed in RCBTB1-deficient human iPSC-derived RPE.13 Additionally, we observed thickening of Bruch's membrane in Rcbtb1-KO mice and rare, drusen-like deposits in the interior of Bruch's membrane, both pathological features previously described in NRF2/peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α) double KO mice, which also show increases in damaged RPE mitochondria.23 Together, these results demonstrate that Rcbtb1-KO mice recapitulate key features of RCBTB1-associated retinopathy, showing a mild retinal dystrophy phenotype with middle-aged onset photoreceptor loss associated with increased oxidative stress and RPE mitochondrial damage.
AAV2–RCBTB1 Gene Replacement Therapy
We previously developed an AAV2-based gene replacement therapy vector for restoring RCBTB1 expression in retinal cells. Treatment of patient iPSC-derived, RCBTB1-deficient RPE cells increased primary cilia growth and partially restored the NRF2 oxidative stress response.12,13 Here, we evaluated the AAV2–RCBTB1 gene therapy vector in Rcbtb1-KO mice.
Subretinal injection of AAV2–RCBTB1 and AAV2–EGFP vectors resulted in long-term expression of RCBTB1 or EGFP in the mouse retina with prominent expression of RCBTB1 immunoreactivity in the inner retinal layers 6 months after treatment. RCBTB1 immunoreactivity was also detected in the RPE cells of AAV2–RCBTB1 treated mice, but not in age-matched WT or untreated Rcbtb1-KO mice. Given that subretinal AAV2 injections primarily transduce photoreceptor and RPE cells, expression of the RCBTB1 transgene in the inner retina suggests that leakage of vector from the subretinal bleb to the vitreous occurred after injection via the trans-retinal route, resulting in transduction of cells in both the inner and outer retina. Rcbtb1-KO mice treated with AAV2–RCBTB1 showed significant increases in outer retinal thickness compared with untreated mice, indicating that RCBTB1 gene replacement prevented outer retinal thinning.
Conclusions
In this study, we present a knockout mouse model of RCBTB1-associated retinopathy. We showed that the loss of Rcbtb1 mRNA and protein expression in knockout mice was associated with the onset of middle-aged outer retinal thinning, accelerated decline of rod-dependent light responses, increased photoreceptor oxidative stress, and RPE mitochondrial damage. AAV2–RCBTB1 vectors were effective in achieving long-term retinal expression of RCBTB1 and preventing outer retinal thinning, providing proof of principle for early intervention gene replacement therapy for RCBTB1-associated retinopathy.
Supplementary Material
Acknowledgments
The authors thank the staff at Ozgene Australia and the Harry Perkins Institute Bioresources Facility for their contributions to the generation and maintenance of mice used in this study.
Supported by the Lee and Low family, the McCusker Charitable Foundation, and a research grant from the National Health and Medical Research Council of Australia (GNT1188694).
Disclosure: S. McLenachan, None; R. Rashwan, None; Z. Huang, None; S.Y. Moon, None; K. Zaw, None; C.R. Hannan, None; S. Pervan, None; D. Zhang, None; L. Griffiths, None; L.S. Carvalho, None; F.K. Chen, None
References
- 1. Catomeris AJ, Ballios BG, Sangermano R, et al.. Novel RCBTB1 variants causing later-onset non-syndromic retinal dystrophy with macular chorioretinal atrophy. Ophthalmic Genet. 2022; 43: 332–339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Coppieters F, Ascari G, Dannhausen K, et al.. Isolated and syndromic retinal dystrophy caused by biallelic mutations in RCBTB1, a gene implicated in ubiquitination. Am J Hum Genet. 2016; 99: 470–480. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Huang Z, Zhang D, Thompson JA, et al.. Deep clinical phenotyping and gene expression analysis in a patient with RCBTB1-associated retinopathy. Ophthalmic Genet. 2021; 42: 266–275. [DOI] [PubMed] [Google Scholar]
- 4. Yang J, Xiao X, Sun W, Li S, Jia X, Zhang Q. Variants in RCBTB1 are associated with autosomal recessive retinitis pigmentosa but not autosomal dominant FEVR. Curr Eye Res. 2021; 46: 839–844. [DOI] [PubMed] [Google Scholar]
- 5. Liu PK, Lee W, Su PY, et al.. Cross-sectional analysis of outer retinal tubulation in inherited retinal diseases: a multicenter study. Am J Ophthalmol. 2025; 269: 116–135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Birtel J, von Landenberg C, Gliem M, et al.. Mitochondrial retinopathy. Ophthalmol Retina. 2022; 6: 65–79. [DOI] [PubMed] [Google Scholar]
- 7. Wu JH, Liu JH, Ko YC, et al.. Haploinsufficiency of RCBTB1 is associated with Coats disease and familial exudative vitreoretinopathy. Hum Mol Genet. 2016; 25: 1637–1647. [DOI] [PubMed] [Google Scholar]
- 8. Holtes LK, de Bruijn SE, Cremers FPM, Roosing S.. Dual inheritance patterns: a spectrum of non-syndromic inherited retinal disease phenotypes with varying molecular mechanisms. Prog Retin Eye Res. 2025; 104: 101308. [DOI] [PubMed] [Google Scholar]
- 9. Mabuchi H, Fujii H, Calin G, et al.. Cloning and characterization of CLLD6, CLLD7, and CLLD8, novel candidate genes for leukemogenesis at chromosome 13q14, a region commonly deleted in B-cell chronic lymphocytic leukemia. Cancer Res. 2001; 61: 2870–2877. [PubMed] [Google Scholar]
- 10. Plafker KS, Singer JD, Plafker SM.. The ubiquitin conjugating enzyme, UbcM2, engages in novel interactions with components of cullin-3 based E3 ligases. Biochemistry. 2009; 48: 3527–3537. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Huang Z, Zhang D, Chen S-C, et al.. Generation of three induced pluripotent stem cell lines from an isolated inherited retinal dystrophy patient with RCBTB1 frameshifting mutations. Stem Cell Res. 2019; 40: 101549. [DOI] [PubMed] [Google Scholar]
- 12. Huang Z, Zhang D, Chen S-C, et al.. Gene replacement therapy restores RCBTB1 expression and cilium length in patient-derived retinal pigment epithelium. J Cell Mol Med. 2021; 25: 10020–10027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Huang Z, Zhang D, Chen S-C, et al.. Mitochondrial dysfunction and impaired antioxidant responses in retinal pigment epithelial cells derived from a patient with RCBTB1-associated retinopathy. Cells. 2023; 12: 1358. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Koentgen F, Lin J, Katidou M, et al.. Exclusive transmission of the embryonic stem cell-derived genome through the mouse germline. Genesis. 2016; 54: 326–333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Westhaus A, Eamegdool SS, Fernando M, et al.. AAV capsid bioengineering in primary human retina models. Sci Rep. 2023; 13: 21946. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Miller AL, Fuller-Carter PI, Masarini K, et al.. Increased H3K27 trimethylation contributes to cone survival in a mouse model of cone dystrophy. Cell Mol Life Sci. 2022; 79: 409. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Hayes MJ, Tracey-White D, Kam JH, Powner MB, Jeffery G.. The 3D organisation of mitochondria in primate photoreceptors. Sci Rep. 2021; 11: 18863. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Ferdous S, Liao KL, Gefke ID, et al.. Age-related retinal changes in wild-type C57BL/6J mice between 2 and 32 months. Invest Ophthalmol Vis Sci. 2021; 62: 9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Carron M, Naert T, Ascari G, et al.. Functional characterization of a Xenopus tropicalis knockout and a human cellular model of RCBTB1-associated inherited retinal disease shows involvement of RCBTB1 in the cellular response to oxidative stress. Invest Ophthalmol Vis Sci. 2020; 61: 1125. [Google Scholar]
- 20. Soubannier V, McLelland GL, Zunino R, et al.. A vesicular transport pathway shuttles cargo from mitochondria to lysosomes. Curr Biol. 2012; 22: 135–141. [DOI] [PubMed] [Google Scholar]
- 21. Soubannier V, Rippstein P, Kaufman BA, Shoubridge EA, McBride HM.. Reconstitution of mitochondria derived vesicle formation demonstrates selective enrichment of oxidized cargo. PLoS One. 2012; 7: e52830. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Sugiura A, McLelland GL, Fon EA, McBride HM.. A new pathway for mitochondrial quality control: mitochondrial-derived vesicles. EMBO J. 2014; 33: 2142–2156. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Felszeghy S, Viiri J, Paterno JJ, et al.. Loss of NRF-2 and PGC-1α genes leads to retinal pigment epithelium damage resembling dry age-related macular degeneration. Redox Biol. 2019; 20: 1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.