This study complied with the Helsinki declaration and was approved by the Ethics Committee of Henan Provincial Eye Hospital for the publication of clinical information, family history and blood extraction for genetic testing [HNEECKY-2019(15)] with the consent of all subjects. Detailed ophthalmic examination including best corrected visual acuity (BCVA), fundus photography, swept source optical coherence tomography (SS-OCT, VG200D SVision Imaging, Henan, China), and full-field electroretinography (ERG, ROLAND CONSULT, Brandenburg, Germany) with skin electrodes. To exclude the possibility that multi-organ was affected, skeletal survey and chest x-ray were then examined. Other members of the family line also underwent BCVA, SS-OCT, and other ocular examinations.

Genomic DNA of subjects was extracted from peripheral blood. Targeted next-generation sequencing (Tg-NGS) of DNA using a custom-designed panel (PS400) which contains 376 known genes involved in inherited retinal diseases, including the exons and adjacent intron regions (50 bp) and known intron mutation (29). The captured DNA was sequenced using a high-throughput sequencer (Illumina) after elution, amplification, and purification. Sequencing data were aligned with the human genome reference (UCSC hg19) using TGex (LifeMap Science, Alameda, United States), Efficient Genosome Intepration System, EGIS (SierraVast Bio-Medical, Shanghai, China), and XY Gene Ranger 2.0 software (Seekgene, Shanghai, China) and genetic variants were identified. Quality parameters such as coverage of target regions and average sequencing depth were collected. The average sequencing depth of the target region for retinal genetic disease-associated gene detection is 200X. Sanger sequencing and family segregation analysis were used to verify suspected disease-associated gene variants in available family members.

SIFT, Polyphen2, LRT, Mutation Taster, CADD, and others were used to predict the pathogenicity of the mutation and gnomAD, ExAC, 1,000 genomes was used to determine allele frequencies of the variants in East Asian populations (30). Clustal Omega and WebLogo were employed for sequence comparison between different species. AphalFold was used to be predicted tertiary structure of protein. Predicting and analyzing changes in the structure and function of mutant proteins were performed by using the PyMoL and HOPE online software. CFAP410 mutant protein stability prediction with the online tool I-Mutant v2.0 and MUpro (31).

The HEK293T cells used in this experiment were purchased from American Type Culture Collection (ATCC, Manassas, VA, United States). Cells were grown in medium containing DMEM of high glucose (SH30022.01, HyClone, UT, United States), 10% fetal bovine serum (35-081-CV, Corning, NY, United States), and 1% penicillin–streptomycin (32105, Mengbio, Chongqing, China). The conditions of the incubator are 37°C, 5% CO₂.

Sangon Biotech (Shanghai, China) was commissioned to synthesize CDS of CFAP410 (NM_004928.3) wild-type, CFAP410 c. 319 T > C and CFAP410 c.347C > T containing a C-terminal Flag and clone into the vector separately. The constructed vectors were confirmed by Sanger sequencing (Sangon Biotech). Well-grown cells with a density of 80% were transfected with EZ Cell Transfection Regent (Life-iLab, Shanghai, China) according to the manufacturer’s instructions.

Cells were lysed with RIPA (PC101, EpiZyme, Shanghai, China), supplemented with 1% protease inhibitor cocktail (GRF101, EpiZyme). Cell lysates were denatured and separated by 12.5% SDS-PAGE electrophoresis, and transferred to polyvinylidene fluoride membranes (IPFL00010, Millipore, MA, United States). The membranes were blocked with 5% skim milk and incubated with primary antibodies mouse anti-Flag (1:5,000; 2367S, Cell Signaling Technology, MA, United States) and rabbit anti-βactin (1:5,000; 4970S, Cell Signaling Technology) at 4°C overnight. After washing with PBS containing 0.1% Tween-20 (PBST), membranes were incubated with secondary antibodies at room temperature for 2 h and visualized with Chemiluminescent detection reagent (WBKLS0500, Millipore). ImageJ software (Version 1.52a, NIH) was used to analyze bands.

After 24 h cell culture, HEK293T cells transfected with pcDNA3.1, Flag-CFAP410, Flag-CFAP410 c.319 T > C, and Flag-CFAP410 c.347C > T were treated with 100 μM cycloheximide (CHX, HY-N0901, MedChemExpress, NJ, United States) or CHX mixed inhibitors of proteasome (MG132). Cells were scraped at 0, 6, 12, and 18 h after CHX/MG132 exposure using ice-cold PBS. The total protein was extracted for Western blot. Wild-type and mutant CFAP410 protein levels were detected using flag antibody. β-actin were the loading control (32).

HEK293T cells were transfected with 12 μg of Flag-CFAP410 WT plasmid, Y107H and P116L plasmids in 10 cm plates that were 80% full of cells, and harvested for protein extraction after 24 h. Total protein lysate was extracted by immunoprecipitation buffer (BL509A, Biosharp, Guangzhou, China), and the concentration of the protein was quantified with BCA protein assay kit (P0011, Beyotime, Shanghai, China). 1,000 μg proteins were mixed with 10 μg anti-Flag magnetic beads (HY-K0207, MedChemExpress) and shaken at 4°C on rotor overnight. Beads with absorbed proteins were gathered by Magnetic Stand (HY-K0200, MedChemExpress) and washed three times by IP-buffer containing 1% protease inhibitor. The beads added to 1× loading buffer boiled for 10 min and the supernatant was collected. The analyzed by western blot.

Cells were divided into the following four groups: pcDNA3.1, WT, Y107H, and P116L. The cells were seeded into six-well tissue culture plates (4 × 10⁵ cells/well). After transfection for 24 h, the cells were collected and washed with 1 × PBS. Added pre-cooled ethanol (concentration 75%) while vortexing on a turbo shaker and refrigerate for more than 12 h. Cells washed with PBS and then 0.5 mL PI/RNase Staining Buffer (550,825, BD, NJ, United States) was added and cells were incubated for 30 min at room temperature. The DNA content was detected using flow cytometry (Canto plus, BD). The percentage of cells in the G1 phase, the S phase, and the G2 phase was analyzed.

Cell crawls were fixed with 4% paraformaldehyde for 30 min at room temperature, washed with PBS and permeabilized with 0.5% Triton X-100 (in PBS) for 20 min. Slides were incubated with 5% bovine serum albumin for 2 h at room temperature and then incubated with anti-mouse IgG-FITC antibodies (1:250; abs20004, Absin, Shanghai, China) overnight at 4°C. After being washed three times with PBS, the secondary anti-mouse Alexaflor 594 (1:250; abs20017, Absin) were applied for 2 h at room temperature in the dark. Nuclei were counterstained with DAPI (F6057, Sigma, MO, STL, United States). Luorescence microscopy images were obtained by Zeiss confocal microscope (NLO780, Zeiss, Oberkochen, Germany).

Statistical software GraphPad Prism was used to analyze the experimental data, and one-way ANOVA was used for multiple comparisons. p < 0.05 was considered statistically significant.

A Chinese family was identified and the pedigree was shown in Figure 2A. The proband, a 6-year-old boy, was noted to have had slowly progressive vision loss and narrow visual field in both eyes since 1 year ago. Night blindness was denied by boy’s parents. At presentation, the best-corrected visual acuity was 0.20 (+0.50/−1.25 × 20) of right eye and 0.20 (+1.00/−2.00 × 170) of left eye. The proband’s right and left eye axial length were 22.23 mm and 22.43 mm, respectively. There were no nystagmus or strabismus. Fundus showed macular staphyloma and uneven granular pigment disorder in the periphery of the retina (Figure 3A). Fundus autofluorescence in both eyes showed an enhanced ring-shape fluorescence in the posterior pole under a low fluorescence background and hyperfluorescence in the peripheral retina. SS-OCT displayed thinning and atrophy of the outer retina, missing of ellipsoid zone (EZ) and interdigitation zone (IZ), with only a small amount of EZ remained in the fovea. In addition, macular staphyloma was noted in both eyes (Figure 3B). Full-field ERG revealed severe reduction in scotopic and photopic ERG responses (Figure 3C). Other examinations including pupils and anterior segments were normal. The proband presented no hearing abnormalities. The chest X-ray showed no significant skeletal abnormalities (Figure 4). The boy’s height and weight were in normal range. Cardiac ultrasound showed no significant abnormalities. No other developmental problems were noticed.

None of remaining family members showed abnormalities related to body development, stature, or ocular history (Supplementary Figure 1).

A total of 12 variants in 11 genes were detected in the blood samples of the proband by Tg-NGS. Except for CFAP410, which contained two heterozygous missense pathogenic variants, only one variant was detected in the remaining 10 genes, and the pathogenicity was unknown. Most of the unrelated variants were excluded (Supplementary Table 1). Two heterozygous missense pathogenic variants, c.319 T > C (p.Tyr107His) and c.347C > T (p.Pro116Leu), in exon 4 of the CFAP410 located at chr.21–45,752,942 and chr.21–45,752,970, were considered to be phenotypically related. Sanger sequencing revealed that the variants segregated in participating individuals and followed an autosomal recessive heritability pattern (Figure 2B).

Y107H was the relatively common pathogenic variant reported in CFAP410, and its pathogenicity has also been verified in vitro. Functional experiments showed that this variant could lead to reduction of protein expression and localization of proteins in cells (16). Recently, homozygous variant Y107H was found in two patients with RP and ALS (33). Y107C, tyrosine changed into cysteine at position 107, has been also reported to cause isolated RP or CORD in homozygous patients (12, 16). According to ACMG guidelines, CFAP410 c.319 T > C (p.Tyr107His) was “pathogenic” (PS1 + PS3 + PM1 + PP1 + PP3).

Another missense pathogenic variant, P116L, was also located in the same LRRCT (leucine-rich repeat C-terminal) structural domain as Y107H and was highly conserved in interspecies amino acid sequence comparisons (Figure 2C). Mutation of a 100% conserved residue is usually damaging for the protein. The P116L was predicted to be harmful by several in silico tools (Supplementary Table 2). According to ACMG guidelines, this variant was “likely pathogenic” (PS3 + PM1 + PP1 + PP3).

The exact 3D-structure of CFAP410 was unknown. We predicted protein model by AlphaFold (34, 35) and the predicted structural changes of the mutant proteins were shown in Figure 5. To predict the effects of pathogenic variants on protein structure and function, we used the online software HOPE. Compared with the wild-type residue, the mutant residue at position 107 was smaller and more hydrophilic (Figure 5B). This change might cause loss of hydrophobic interactions with other molecules on the surface of the protein. The variant residue at position 116 was bigger, which might lead to bumps (Figure 5B). LRRCT is critical for the folding of proteins (36), thus alternations in sizes and properties of amino acid may lead to changes in stability. Indeed, results obtained from the online tool I-Mutant v2.0 suggested that the stability of the mutant CFAP410 proteins decreased. However, prediction software MUpro showed increased stability.

Ciliogenesis is closely related to the cell cycle. It was shown that NEK1, a causative gene of ciliopathy, was required in regulating cell cycle (37). CFAP410 and NEK1 belonged to the ciliary functional modules and played a consistent role in ciliogenesis.

Therefore, we verified the effect of mutant proteins on cell cycle. We constructed wild-type (CFAP410-WT), mutant (CFAP410-P116L and CFAP410-Y107H) plasmids using pcDNA3.1 as vectors and transfected them into HEK293T cells. 24 h after transfection with plasmids into cultured cells. G1 phase were blocked in the cell cycle in the two pathogenic variant groups when compared with the empty vector (pcDNA3.1) and wild-type group (Figure 6).

Pathogenic variant located in the LRRCT may affect the stability of the protein and I-Mutant v2.0 online tool predicted reduction in the stability of the mutant proteins.

To determine changes in mutant protein content, immunofluorescence experiments were performed. Fluorescence intensity of the mutants Y107H and P116L transferred cells was significantly weaker than that of the wild type (Figure 7A). We found that the mutant proteins were unevenly distributed in the cytoplasm and some of the proteins were presented in clusters.

We then performed CHX assay to verify the changes in protein stability. At 18 h after plasmid transfection, HEK293T cells in each group were treated with 100 μM CHX for 0, 6, 12, and 18 h. Western blot showed that the half-life of CFAP410 protein carrying the Y107H variant and P116 L variant was significantly shorter compared with the wild-type protein, indicating that both pathogenic variants impaired the stability of CFAP410 protein (Figure 7B). The ubiquitin-proteasome system is essential for maintaining protein homeostasis at the level of protein degradation. We added MG132, a protease inhibitor, to CHX-treated cells. The results showed a significant increase in protein stability of the CFAP410 mutants transferred cells (Figure 7C). Subsequently, we examined the ubiquitination levels of CFAP410 proteins in HEK293T cells. We found that the ubiquitination levels of P116L and Y107H proteins were significantly increased compared to the wild-type CFAP410 proteins (Figure 7D). The data suggested that the reduced stability of the two mutant proteins might be associated with the ubiquitin-proteasome pathway.