A Long-Term Retrospective Natural History Study of EFEMP1-Associated Autosomal Dominant Drusen.

Purpose: To analyze the natural history of EFEMP1-associated autosomal dominant drusen (ADD). Methods: In this retrospective observational study of molecularly confirmed patients with ADD, data and retinal imaging were extracted from an in-house database. The main outcome measurements were best-corrected visual acuity (BCVA), refraction, and retinal imaging, including quantitative analyses of the outer nuclear layer (ONL) thickness and pigmented epithelium detachment area, as well as qualitative analyses. Results: The study included 44 patients (34 females and 10 males). The mean ± SD age of symptom onset was 40.1 ± 6.59 years of age (range, 25-52). Fourteen patients were asymptomatic during their entire follow-up. The most common symptoms at presentation were reduced vision (70%) and distortion in central vision (53%). Most subjects were emmetropic. The mean BCVA (logMAR) at baseline was 0.27 ± 0.41 (range, -0.1 to 2.1) in right eyes and was 0.19 ± 0.32 (range, -0.2 to 1.3) in left eyes. After a mean follow-up of 7.9 years, BCVA was reduced to 0.59 ± 0.66 (range, -0.1 to 2.1) in right eyes and 0.5 ± 0.72 (range, -0.1 to 2.4) in left eyes, values that were significantly different than baseline (P < 0.0001 and P < 0.0014, respectively). Fifteen eyes showed active or inactive choroidal neovascularization (CNV). BCVA differed significantly (P = 0.0004) between eyes with and without CNV at a comparable mean age. The ONL had a slow rate of thinning longitudinally, which significantly correlated with BCVA. Conclusions: Despite the late onset and relatively good prognosis of ADD, CNVs are more frequent than previously reported and are associated with a worse prognosis. Further research is necessary to elucidate gender associations.

Effective AAV-mediated gene replacement therapy in retinal organoids modeling AIPL1-associated LCA4.

Biallelic variations in the aryl hydrocarbon receptor interacting protein-like 1 (AIPL1) gene cause Leber congenital amaurosis subtype 4 (LCA4), an autosomal recessive early-onset severe retinal dystrophy that leads to the rapid degeneration of retinal photoreceptors and the severe impairment of sight within the first few years of life. Currently, there is no treatment or cure for AIPL1-associated LCA4. In this study, we investigated the potential of adeno-associated virus-mediated AIPL1 gene replacement therapy in two previously validated human retinal organoid (RO) models of LCA4. We report here that photoreceptor-specific AIPL1 gene replacement therapy, currently being tested in a first-in-human application, effectively rescued molecular features of AIPL1-associated LCA4 in these models. Notably, the loss of retinal phosphodiesterase 6 was rescued and elevated cyclic guanosine monophosphate (cGMP) levels were reduced following treatment. Transcriptomic analysis of untreated and AAV-transduced ROs revealed transcriptomic changes in response to elevated cGMP levels and viral infection, respectively. Overall, this study supports AIPL1 gene therapy as a promising therapeutic intervention for LCA4.

CRISPR-Cas9 correction of a nonsense mutation in LCA5 rescues lebercilin expression and localization in human retinal organoids.

Mutations in the lebercilin-encoding gene LCA5 cause one of the most severe forms of Leber congenital amaurosis, an early-onset retinal disease that results in severe visual impairment. Here, we report on the generation of a patient-specific cellular model to study LCA5-associated retinal disease. CRISPR-Cas9 technology was used to correct a homozygous nonsense variant in LCA5 (c.835C>T; p.Q279∗) in patient-derived induced pluripotent stem cells (iPSCs). The absence of off-target editing in gene-corrected (isogenic) control iPSCs was demonstrated by whole-genome sequencing. We differentiated the patient, gene-corrected, and unrelated control iPSCs into three-dimensional retina-like cells, so-called retinal organoids. We observed opsin and rhodopsin mislocalization to the outer nuclear layer in patient-derived but not in the gene-corrected or unrelated control organoids. We also confirmed the rescue of lebercilin expression and localization along the ciliary axoneme within the gene-corrected organoids. Here, we show the potential of combining precise single-nucleotide gene editing with the iPSC-derived retinal organoid system for the generation of a cellular model of early-onset retinal disease.

Eupatilin Improves Cilia Defects in Human CEP290 Ciliopathy Models.

The photoreceptor outer segment is a highly specialized primary cilium that is essential for phototransduction and vision. Biallelic pathogenic variants in the cilia-associated gene CEP290 cause non-syndromic Leber congenital amaurosis 10 (LCA10) and syndromic diseases, where the retina is also affected. While RNA antisense oligonucleotides and gene editing are potential treatment options for the common deep intronic variant c.2991+1655A>G in CEP290, there is a need for variant-independent approaches that could be applied to a broader spectrum of ciliopathies. Here, we generated several distinct human models of CEP290-related retinal disease and investigated the effects of the flavonoid eupatilin as a potential treatment. Eupatilin improved cilium formation and length in CEP290 LCA10 patient-derived fibroblasts, in gene-edited CEP290 knockout (CEP290 KO) RPE1 cells, and in both CEP290 LCA10 and CEP290 KO iPSCs-derived retinal organoids. Furthermore, eupatilin reduced rhodopsin retention in the outer nuclear layer of CEP290 LCA10 retinal organoids. Eupatilin altered gene transcription in retinal organoids by modulating the expression of rhodopsin and by targeting cilia and synaptic plasticity pathways. This work sheds light on the mechanism of action of eupatilin and supports its potential as a variant-independent approach for CEP290-associated ciliopathies.

Retinal Organoids from an AIPL1 CRISPR/Cas9 Knockout Cell Line Successfully Recapitulate the Molecular Features of LCA4 Disease.

Aryl hydrocarbon receptor-interacting protein-like 1 (AIPL1) is expressed in photoreceptors where it facilitates the assembly of phosphodiesterase 6 (PDE6) which hydrolyses cGMP within the phototransduction cascade. Genetic variations in AIPL1 cause type 4 Leber congenital amaurosis (LCA4), which presents as rapid loss of vision in early childhood. Limited in vitro LCA4 models are available, and these rely on patient-derived cells harbouring patient-specific AIPL1 mutations. While valuable, the use and scalability of individual patient-derived LCA4 models may be limited by ethical considerations, access to patient samples and prohibitive costs. To model the functional consequences of patient-independent AIPL1 mutations, CRISPR/Cas9 was implemented to produce an isogenic induced pluripotent stem cell line harbouring a frameshift mutation in the first exon of AIPL1. Retinal organoids were generated using these cells, which retained AIPL1 gene transcription, but AIPL1 protein was undetectable. AIPL1 knockout resulted in a decrease in rod photoreceptor-specific PDE6α and ß, and increased cGMP levels, suggesting downstream dysregulation of the phototransduction cascade. The retinal model described here provides a novel platform to assess functional consequences of AIPL1 silencing and measure the rescue of molecular features by potential therapeutic approaches targeting mutation-independent pathogenesis.

HSP70-HSP90 Chaperone Networking in Protein-Misfolding Disease.

Molecular chaperones and their associated co-chaperones are essential in health and disease as they are key facilitators of protein-folding, quality control and function. In particular, the heat-shock protein (HSP) 70 and HSP90 molecular chaperone networks have been associated with neurodegenerative diseases caused by aberrant protein-folding. The pathogenesis of these disorders usually includes the formation of deposits of misfolded, aggregated protein. HSP70 and HSP90, plus their co-chaperones, have been recognised as potent modulators of misfolded protein toxicity, inclusion formation and cell survival in cellular and animal models of neurodegenerative disease. Moreover, these chaperone machines function not only in folding but also in proteasome-mediated degradation of neurodegenerative disease proteins. This chapter gives an overview of the HSP70 and HSP90 chaperones, and their respective regulatory co-chaperones, and explores how the HSP70 and HSP90 chaperone systems form a larger functional network and its relevance to counteracting neurodegenerative disease associated with misfolded proteins and disruption of proteostasis.

Induced Pluripotent Stem Cells and Genome-Editing Tools in Determining Gene Function and Therapy for Inherited Retinal Disorders.

Inherited retinal disorders (IRDs) affect millions of people worldwide and are a major cause of irreversible blindness. Therapies based on drugs, gene augmentation or transplantation approaches have been widely investigated and proposed. Among gene therapies for retinal degenerative diseases, the fast-evolving genome-editing CRISPR/Cas technology has emerged as a new potential treatment. The CRISPR/Cas system has been developed as a powerful genome-editing tool in ophthalmic studies and has been applied not only to gain proof of principle for gene therapies in vivo, but has also been extensively used in basic research to model diseases-in-a-dish. Indeed, the CRISPR/Cas technology has been exploited to genetically modify human induced pluripotent stem cells (iPSCs) to model retinal disorders in vitro, to test in vitro drugs and therapies and to provide a cell source for autologous transplantation. In this review, we will focus on the technological advances in iPSC-based cellular reprogramming and gene editing technologies to create human in vitro models that accurately recapitulate IRD mechanisms towards the development of treatments for retinal degenerative diseases.

Investigation of PTC124-mediated translational readthrough in a retinal organoid model of AIPL1-associated Leber congenital amaurosis.

Leber congenital amaurosis type 4 (LCA4), caused by AIPL1 mutations, is characterized by severe sight impairment in infancy and rapidly progressing degeneration of photoreceptor cells. We generated retinal organoids using induced pluripotent stem cells (iPSCs) from renal epithelial cells obtained from four children with AIPL1 nonsense mutations. iPSC-derived photoreceptors exhibited the molecular hallmarks of LCA4, including undetectable AIPL1 and rod cyclic guanosine monophosphate (cGMP) phosphodiesterase (PDE6) compared with control or CRISPR-corrected organoids. Increased levels of cGMP were detected. The translational readthrough-inducing drug (TRID) PTC124 was investigated as a potential therapeutic agent. LCA4 retinal organoids exhibited low levels of rescue of full-length AIPL1. However, this was insufficient to fully restore PDE6 in photoreceptors and reduce cGMP. LCA4 retinal organoids are a valuable platform for in vitro investigation of novel therapeutic agents.

The Role of Hsp90 in Retinal Proteostasis and Disease.

Photoreceptors are sensitive neuronal cells with great metabolic demands, as they are responsible for carrying out visual phototransduction, a complex and multistep process that requires the exquisite coordination of a large number of signalling protein components. Therefore, the viability of photoreceptors relies on mechanisms that ensure a well-balanced and functional proteome that maintains the protein homeostasis, or proteostasis, of the cell. This review explores how the different isoforms of Hsp90, including the cytosolic Hsp90α/ß, the mitochondrial TRAP1, and the ER-specific GRP94, are involved in the different proteostatic mechanisms of photoreceptors, and elaborates on Hsp90 function when retinal homeostasis is disturbed. In addition, several studies have shown that chemical manipulation of Hsp90 has significant consequences, both in healthy and degenerating retinae, and this can be partially attributed to the fact that Hsp90 interacts with important photoreceptor-associated client proteins. Here, the interaction of Hsp90 with the retina-specific client proteins PDE6 and GRK1 will be further discussed, providing additional insights for the role of Hsp90 in retinal disease.

CRISPR-Cas9 correction of OPA1 c.1334G>A: p.R445H restores mitochondrial homeostasis in dominant optic atrophy patient-derived iPSCs.

Autosomal dominant optic atrophy (DOA) is the most common inherited optic neuropathy in the United Kingdom. DOA has an insidious onset in early childhood, typically presenting with bilateral, central visual loss caused by the preferential loss of retinal ganglion cells. 60%-70% of genetically confirmed DOA cases are associated with variants in OPA1, a ubiquitously expressed GTPase that regulates mitochondrial homeostasis through coordination of inner membrane fusion, maintenance of cristae structure, and regulation of bioenergetic output. Whether genetic correction of OPA1 pathogenic variants can alleviate disease-associated phenotypes remains unknown. Here, we demonstrate generation of patient-derived OPA1 c.1334G>A: p.R445H mutant induced pluripotent stem cells (iPSCs), followed by correction of OPA1 through CRISPR-Cas9-guided homology-directed repair (HDR) and evaluate the effect of OPA1 correction on mitochondrial homeostasis. CRISPR-Cas9 gene editing demonstrated an efficient method of OPA1 correction, with successful gene correction in 57% of isolated iPSCs. Correction of OPA1 restored mitochondrial homeostasis, re-establishing the mitochondrial network and basal respiration and ATP production levels. In addition, correction of OPA1 re-established the levels of wild-type (WT) mitochondrial DNA (mtDNA) and reduced susceptibility to apoptotic stimuli. These data demonstrate that nuclear gene correction can restore mitochondrial homeostasis and improve mtDNA integrity in DOA patient-derived cells carrying an OPA1 variant.

Clinical and functional analyses of AIPL1 variants reveal mechanisms of pathogenicity linked to different forms of retinal degeneration.

Disease-causing sequence variants in the highly polymorphic AIPL1 gene are associated with a broad spectrum of inherited retinal diseases ranging from severe autosomal recessive Leber congenital amaurosis to later onset retinitis pigmentosa. AIPL1 is a photoreceptor-specific co-chaperone that interacts with HSP90 to facilitate the stable assembly of retinal cGMP phosphodiesterase, PDE6. In this report, we establish unequivocal correlations between patient clinical phenotypes and in vitro functional assays of uncharacterized AIPL1 variants. We confirm that missense and nonsense variants in the FKBP-like and tetratricopeptide repeat domains of AIPL1 lead to the loss of both HSP90 interaction and PDE6 activity, confirming these variants cause LCA. In contrast, we report the association of p.G122R with milder forms of retinal degeneration, and show that while p.G122R had no effect on HSP90 binding, the modulation of PDE6 cGMP levels was impaired. The clinical history of these patients together with our functional assays suggest that the p.G122R variant is a rare hypomorphic allele with a later disease onset, amenable to therapeutic intervention. Finally, we report the primate-specific proline-rich domain to be dispensable for both HSP90 interaction and PDE6 activity. We conclude that variants investigated in this domain do not cause disease, with the exception of p.A352_P355del associated with autosomal dominant cone-rod dystrophy.

The ubiquitin-like modifier FAT10 inhibits retinal PDE6 activity and mediates its proteasomal degradation.

The retina-specific chaperone aryl hydrocarbon interacting protein-like 1 (AIPL1) is essential for the correct assembly of phosphodiesterase 6 (PDE6), which is a pivotal effector enzyme for phototransduction and vision because it hydrolyzes cGMP. AIPL1 interacts with the cytokine-inducible ubiquitin-like modifier FAT10, which gets covalently conjugated to hundreds of proteins and targets its conjugation substrates for proteasomal degradation, but whether FAT10 affects PDE6 function or turnover is unknown. Here, we show that FAT10 mRNA is expressed in human retina and identify rod PDE6 as a retina-specific substrate of FAT10 conjugation. We found that AIPL1 stabilizes the FAT10 monomer and the PDE6-FAT10 conjugate. Additionally, we elucidated the functional consequences of PDE6 FAT10ylation. On the one hand, we demonstrate that FAT10 targets PDE6 for proteasomal degradation by formation of a covalent isopeptide linkage. On the other hand, FAT10 inhibits PDE6 cGMP hydrolyzing activity by noncovalently interacting with the PDE6 GAFa and catalytic domains. Therefore, FAT10 may contribute to loss of PDE6 and, as a consequence, degeneration of retinal cells in eye diseases linked to inflammation and inherited blindness-causing mutations in AIPL1.

Negative Regulator of Ubiquitin-Like Protein 1 modulates the autophagy-lysosomal pathway via p62 to facilitate the extracellular release of tau following proteasome impairment.

Negative regulator of ubiquitin-like protein 1 (NUB1) and its longer isoform NUB1L are ubiquitin-like (UBL)/ubiquitin-associated (UBA) proteins that facilitate the targeting of proteasomal substrates, including tau, synphilin-1 and huntingtin. Previous data revealed that NUB1 also mediated a reduction in tau phosphorylation and aggregation following proteasome inhibition, suggesting a switch in NUB1 function from targeted proteasomal degradation to a role in autophagy. Here, we delineate the mechanisms of this switch and show that NUB1 interacted specifically with p62 and induced an increase in p62 levels in a manner facilitated by inhibition of the proteasome. NUB1 moreover increased autophagosomes and the recruitment of lysosomes to aggresomes following proteasome inhibition. Autophagy flux assays revealed that NUB1 affected the autophagy-lysosomal pathway primarily via the UBA domain. NUB1 localized to cytosolic inclusions with pathological forms of tau, as well as LAMP1 and p62 in the hippocampal neurons of tauopathy mice. Finally, NUB1 facilitated the extracellular release of tau following proteasome inhibition. This study thus shows that NUB1 plays a role in regulating the autophagy-lysosomal pathway when the ubiquitin proteasome system is compromised, thus contributing to the mechanisms targeting the removal of aggregation-prone proteins upon proteasomal impairment.

Gene and Cell Therapy for AIPL1-Associated Leber Congenital Amaurosis: Challenges and Prospects.

Leber congenital amaurosis (LCA) caused by AIPL1 mutations is one of the most severe forms of inherited retinal degeneration (IRD). The rapid and extensive photoreceptor degeneration challenges the development of potential treatments. Nevertheless, preclinical studies show that both gene augmentation and photoreceptor transplantation can regenerate and restore retinal function in animal models of AIPL1-associated LCA. However, questions regarding long-term benefit and safety still remain as these therapies advance towards clinical application. Ground-breaking advances in stem cell technology and genome editing are examples of alternative therapeutic approaches and address some of the limitations associated with previous methods. The continuous development of these cutting-edge biotechnologies paves the way towards a bright future not only for AIPL1-associated LCA patients but also other forms of IRD.

The Leber Congenital Amaurosis-Linked Protein AIPL1 and Its Critical Role in Photoreceptors.

Mutations in the photoreceptor/pineal-expressed gene, aryl hydrocarbon receptor-interacting protein-like 1 (AIPL1), are mainly associated with autosomal recessive Leber congenital amaurosis (LCA), the most severe form of inherited retinopathy that occurs in early childhood. AIPL1 functions as a photoreceptor-specific molecular co-chaperone that interacts specifically with the molecular chaperones HSP90 and HSP70 to facilitate the correct folding and assembly of the retinal cGMP phosphodiesterase (PDE6) holoenzyme. The absence of AIPL1 leads to a dramatic degeneration of rod and cone cells and a complete loss of any light-dependent electrical response. Here we review the important role of AIPL1 in photoreceptor functionality.

Primate retinal cones express phosphorylated tau associated with neuronal degeneration yet survive in old age.

Photoreceptor cells have high energy demands and suffer significantly with age. In aged rodents both rods and cones are lost, but in primates there is no evidence for aged cone loss, although their function declines. Here we ask if aged primate cones suffer from reduced function because of declining metabolic ability. Tau is a microtubule associated protein critical for mitochondrial function in neurons. Its phosphorylation is a feature of neuronal degeneration undermining respiration and mitochondrial dynamics. We show that total tau is widely distributed in the primate outer retina with little age-related change, being present in both rods and cones and their processes. However, all cones specifically accumulate phosphorylated tau, which was not seen in rods. The presence of this protein will likely undermine cone cell function. However, tau phosphorylation inhibits apoptosis. These data may explain why aged primate cones have reduced function but appear to be resistant to cell death. Consequently, therapies designed to remove phosphorylated tau may carry the risk of inducing cone photoreceptor cell death and further undermine ageing visual function.

The integrity and organization of the human AIPL1 functional domains is critical for its role as a HSP90-dependent co-chaperone for rod PDE6.

Biallelic mutations in the photoreceptor-expressed aryl hydrocarbon receptor interacting protein-like 1 (AIPL1) are associated with autosomal recessive Leber congenital amaurosis (LCA), the most severe form of inherited retinopathy in early childhood. AIPL1 functions as a photoreceptor-specific co-chaperone that interacts with the molecular chaperone HSP90 to facilitate the stable assembly of the retinal cyclic GMP (cGMP) phosphodiesterase (PDE6) holoenzyme. In this study, we characterized the functional deficits of AIPL1 variations, some of which induce aberrant pre-mRNA AIPL1 splicing leading to the production of alternative AIPL1 isoforms. We investigated the ability of the AIPL1 variants to mediate an interaction with HSP90 and modulate the rod cGMP PDE6 stability and activity. Our data revealed that both the FK506 binding protein (FKBP)-like domain and the tetratricopeptide repeat (TPR) domain of AIPL1 are required for interaction with HSP90. We further demonstrate that AIPL1 significantly modulates the catalytic activity of heterologously expressed rod PDE6. Although the N-terminal FKBP-like domain of AIPL1 binds the farnesylated PDE6α subunit through direct interaction with the farnesyl moiety, mutations compromising the integrity of the C-terminal TPR domain of AIPL1 also failed to modulate PDE6 activity efficiently. These AIPL1 variants moreover failed to promote the HSP90-dependent stabilization of the PDE6α subunit in the cytosol. In summary, we have successfully validated the disease-causing status of the AIPL1 variations in vitro. Our findings provide insight into the mechanism underlying the co-chaperone role of AIPL1 and will be critical for ensuring an early and effective diagnosis of AIPL1 LCA patients.

Investigation of Aberrant Splicing Induced by AIPL1 Variations as a Cause of Leber Congenital Amaurosis.

PURPOSE: Biallelic mutations in AIPL1 cause Leber congenital amaurosis (LCA), a devastating retinal degeneration characterized by the loss or severe impairment of vision within the first few years of life. AIPL1 is highly polymorphic with more than 50 mutations and many more polymorphisms of uncertain pathogenicity identified. As such, it can be difficult to assign disease association of AIPL1 variations. In this study, we investigate suspected disease-associated AIPL1 variations, including nonsynonymous missense and intronic variants to validate their pathogenicity. METHODS: AIPL1 minigenes harboring missense and intronic variations were constructed by amplification of genomic fragments of the human AIPL1 gene. In vitro splice assays were performed to identify the resultant AIPL1 transcripts. RESULTS: We show that all nine of the suspected disease-associated AIPL1 variations investigated induced aberrant pre-mRNA splicing of the AIPL1 gene, and our study is the first to show that AIPL1 missense mutations alter AIPL1 splicing. We reveal that the presumed rare benign variant c.784G>A [p.(G262S)] alters in vitro AIPL1 splicing, thereby validating the disease-association and clarifying the underlying disease mechanism. We also reveal that in-phase exon skipping occurs normally at a low frequency in the retina, but arises abundantly as a consequence of specific AIPL1 variations, suggesting a tolerance threshold for the expression of these alternative transcripts in the retina normally, which is exceeded in LCA. CONCLUSIONS: Our data confirm the disease-association of the AIPL1 variations investigated and reveal for the first time that aberrant splicing of AIPL1 is an underlying mechanism of disease in LCA.

The Leber congenital amaurosis protein AIPL1 and EB proteins co-localize at the photoreceptor cilium.

PURPOSE: The aim of this study was to investigate the interaction and co-localization of novel interacting proteins with the Leber congenital amaurosis (LCA) associated protein aryl hydrocarbon receptor interacting protein-like 1 (AIPL1). METHODS: The CytoTrapXR yeast two-hybrid system was used to screen a bovine retinal cDNA library. A novel interaction between AIPL1 and members of the family of EB proteins was confirmed by directed yeast two-hybrid analysis and co-immunoprecipitation assays. The localization of AIPL1 and the EB proteins in cultured cells and in retinal cryosections was examined by immunofluorescence microscopy and cryo-immunogold electron microscopy. RESULTS: Yeast two-hybrid (Y2H) analysis identified the interaction between AIPL1 and the EB proteins, EB1 and EB3. EB1 and EB3 were specifically co-immunoprecipitated with AIPL1 from SK-N-SH neuroblastoma cells. In directed 1:1 Y2H analysis, the interaction of EB1 with AIPL1 harbouring the LCA-causing mutations A197P, C239R and W278X was severely compromised. Immunofluorescent confocal microscopy revealed that AIPL1 did not co-localize with endogenous EB1 at the tips of microtubules, endogenous EB1 at the microtubule organising centre following disruption of the microtubule network, or with endogenous ß-tubulin. Moreover, AIPL1 did not localize to primary cilia in ARPE-19 cells, whereas EB1 co-localized with the centrosomal marker pericentrin at the base of primary cilia. However, both AIPL1 and the EB proteins, EB1 and EB3, co-localized with centrin-3 in the connecting cilium of photoreceptor cells. Cryo-immunogold electron microscopy confirmed the co-localization of AIPL1 and EB1 in the connecting cilia in human retinal photoreceptors. CONCLUSIONS: AIPL1 and the EB proteins, EB1 and EB3, localize at the connecting cilia of retinal photoreceptor cells, but do not co-localize in the cellular microtubule network or in primary cilia in non-retinal cells. These findings suggest that AIPL1 function in these cells is not related to the role of EB proteins in microtubule dynamics or primary ciliogenesis, but that their association may be related to a specific role in the specialized cilia apparatus of retinal photoreceptors.