Ocular Characteristics of Patients with Leber Congenital Amaurosis 6 Caused by Pathogenic RPGRIP1 Gene Variation in a Chinese Cohort

Purpose To delineate the clinical and genetic characteristics of Chinese patients with RPGRIP1-associated Leber congenital amaurosis 6 (LCA6). Methods After screening 352 unrelated families with clinically diagnosed RP, five LCA6 patients with RPGRIP1 variations from unrelated Chinese families were identified. Full ophthalmology examinations, including decimal best-corrected visual acuity (BCVA), fundus photography, fundus autofluorescence imaging, spectral-domain optical coherence tomography (SD-OCT), full-field electroretinography (ffERG), multifocal electroretinography (mfERG), perimetry, and flash visual evoked potential (FVEP), were performed. Target next-generation sequencing (NGS) and Sanger sequencing were performed for the five patients to identify and to validate candidate disease-causing variants. Results Five patients were molecularly diagnosed as the LCA6 associated with RPGRIP1 variation, with typical clinical characteristics including congenital night blindness, nystagmus, and visual defect, at an early age. Interestingly, LCA6 exhibited extensive clinical heterogeneity and the changes in the morphology and function were not completely consistent in the five LCA6 patients. Case 1 showed extensive inferior-nasal retinal atrophy with a corresponding area of hypofluorescence in fundus autofluorescence, and the fundus photograph was nearly normal in cases 2 and 3. The ERG results displayed a moderately reduced rod-system response in cases 1 and 2 and a significant reduced rod-system response in case 3. Both case 4 and case 5 showed mottled pigmentation in fundi and an unrecordable rod and cone-system response in ERG. Moreover, we identified eight compound variants and one homozygous variant in the five patients with RPGRIP1. Conclusions This is the largest report focused on the clinical electrophysiological features of patients with associated LCA6 caused by the variation in the RPGRIP1 gene in the Chinese population with an enriched phenotypic and genotypic background of LCA6 to improve future gene therapies.

Among them, Leber congenital amaurosis 6 (LCA6, MIM 613826) is caused by variations in the RPGRIP1 gene encoding retinitis pigmentosa GTPase regulator interacting protein 1 (RPGRIP1), which participates in ciliary transport processes. LCA6 is inherited as an autosomal recessive trait and is estimated to account for 6% of patients with LCA in America [3] and 8.8% of LCA cases in China [4]. LCA6 patients present with a stable and nonprogressive disease course after the initial rapid decline in visual function, and the morphology of photoreceptors in the central retina can persist for a long time. us, LCA6 associated with RPGRIP1 gene variants are expected to be treated by a gene replacement strategy [5,6]; however, despite the latest advances in gene therapy on LCA2 (RPE65) [7] and LCA10 (CEP290) [8], there is no available clinically applicable treatment for LCA6 at present.
It is important to objectively assess the fundus morphology and visual pathway of LCA6 patients as well as the application of a visual function evaluation after treatment in the future. In this study, we aimed to delineate the genotype and phenotype in five unrelated Chinese LCA families with identified RPGRIP1 variants by targeting next-generation sequencing (NGS) technologies.

Patient Recruitment.
Five patients from five unrelated families with a clinical diagnosis of LCA6 were screened from 352 unrelated families with inherited retinal degeneration at the Ophthalmology Department, Southwest Hospital, Army Medical University, Chongqing, China, from 2014 to 2018 and were retrospectively included in this study [9]. Informed consent was obtained from each subject prior to proceeding with the examinations. e study was approved by the local ethics committee (reference number: KY2020096), and all procedures were performed in accordance with the principles of the Declaration of Helsinki. A detailed family history was obtained through interviews with the patients and their relatives.

Molecular Genetic Analysis
Using NGS. Genomic DNA extraction, targeted next-generation sequencing, and bioinformatics analysis were followed by the methods of Dr. Meng et al. 2021 [10]. Targeted next-generation sequencing (NGS) was performed using a capture panel including 195 known inherited retinal disease (IRD) genes (Supplementary Table 1). e clinical phenotypic findings and cosegregation analysis (by Sanger sequencing) were fully considered when we identified the disease-causing variants of patients.

Clinical Findings.
Five affected subjects from five unrelated families with a diagnosis of LCA6 and who harbored RPGRIP1 variants were identified ( Figure 1). e clinical features are summarized in Table 1.
e patients were all female and exhibited congenital night blindness, nystagmus, and severe visual impairment in their infancies. e age at disease diagnosis ranged from 5 to 34 years, and the clinical course of LCA6 displayed a different sequence of events among the five probands. Case 4 and case 5 had the worst visual acuity, only light perception, and HM. eir fundus appearances were also similar, and they exhibited retinal atrophy with intraretinal vascular attenuation, a waxy disc, and the pigment mottling pattern (Figures 2(d) and 2(q)). Accordingly, the fundus autofluorescence revealed an area of a hyper-autofluorescence ring in the para-foveal region in case 4 (Figures 2(h)), and her SD-OCT images (Figures 2(l) and 2(p)) revealed a worsened preservation of adjacent lamellar structures in which retinal pigment epithelium (RPE) and photoreceptor cells showed different degrees of loss in the central macula. Moreover, the visual field of these two patients was too poor to measure and the ERG responses of DA and LA were unrecordable.
e BVCA of patient case 3 was only 0.05 in both eyes, the hypofluorescence area of the macula was enlarged, with a hyperfluorescent ring around the optic disc and the subtemporal mottled hyperfluorescence, and her fundus photographs and were similar to case 2. Case 1 and case 2 showed both 0.15 OD and 0.15 OS. e fundus photographs were grossly normal in case 2 (Figures 2(b) and 2(f )) and case 3 (Figures 2(c) and 2(g)), while case 1 exhibited mottled pigmentary and depigmentation, vascular attenuation, and extensive inferior-nasal retinal atrophy with a corresponding area of hypofluorescence (Figures 2(a) and 2(e)). e SD-OCT images revealed a thinner photoreceptor outer nuclear layer (ONL) with a preserved ellipsoid zone in case 2 (Figures 2(j) and 2(n)) and case 3 (Figures (2(i), 2(k), 2(m), and 2(o)). In addition, the horizontal length of the remaining ellipsoid zone of case 1 was about 1579 μm in the right eye and 1805 μm in the left eye (Figures (2(i), 2(k), 2(m), and 2(o)). e ellipsoid zones of case 2 and case 3 were observed at a horizontal length of 6 mm. e visual field test indicated symmetry defects in superior-temporal vision fields as observed on the hypofluorescence in the right eye of case 1 (Supplemental Figure 1A), a tubular visual field in case 3 (Supplemental figure 1C), and a small center visual island defect in case 2 (Supplemental figure 1B). e visual fields of the remaining two patients were too poor to measure. e mfERG revealed that the density of the wave amplitude had significantly declined in the right eyes of cases 1 and 2 (Supplemental Figure 1D and 1E) and the densities of the other three were too poor to measure. e ERG amplitudes of the a-and b-waves of patients with LCA6 were different compared with the normal control, and the differences were classified into two groups: Group 1 was assessed by the ERG response of DA as more than almost a quarter of the normal control and included case 1 and case 2 ( Figure 3). Group 2 was defined by an ERG response of DA of less than almost a quarter of the normal control and included case 3, case 4, and case 5 ( Figure 3). e response of the cone (LA 3.0 and LA 3.0 flicker ERG) was unrecordable for all five patients ( Figure 3). Case 5 showed a severely reduced amplitude of all waves in the FVEP response, especially the amplitude of P2, which was worse than case 1 and case 2 (Supplemental Figure 2), but the patient's visual function was still assessed. All the basic symptoms supported the clinical diagnosis of LCA6. However, it is interesting that there was an inconsistency in the morphology and function among case 1, case 2, and case 3. e function was similar, but the morphology was different for case 1 and case 2. Furthermore, the morphology was the same, but the function was different for case 2 and case 3. And, the clinical symptom of case 4 and case 5 was typical LCA6.

Molecular Genetic Findings.
We screened for variants of the five affected patients from five unrelated and nonconsanguineous families with known IRD-associated genes using NGS data (    [11] and p.R1189 * in exon 22 [12] have been previously reported to be pathogenic variants of LCA in the compound heterozygous state, and the remaining seven variants have never been reported in ClinVar or HGMD Professional. Moreover, the allele frequency for four RPGRIP1 variants (p.E179Sfs * 11, p.G746R, c.1151+1G > A, and p.R1189 * ) were extremely low at 0.001441%, 0.0008747%, 0.0004091%, and 0.002007% in gnomAD, respectively. e remaining five RPGRIP1 variants were absent from the controls in the 1000 genome, gnomAD, or EXAC databases. According to the American College of Medical Genetics and Genomics guidelines, the five variants, p.E179Sfs * 11, p.E188 * , c.1151+1G > A, p.R1189 * , and c.1467 + 2T > C, were assessed to be "pathogenic," the variants, p.G746R, p.T652A, and p.D862G, were considered to be "likely pathogenic," and the variant p.Y929C was of "uncertain significance."

Discussion
RPGRIP1 has also been reported to be associated with conerod dystrophy 13 (CROD 13), in which the phenotype of patients seemed to have been characterized by a rapid loss of vision in the second decade of life, but further details were not provided [13]. e clinical distinction between LCA and CORD is not clear, and there are large overlaps in the phenotype. In the present study, we reported the detailed clinical and genetic characteristics of five unrelated patients who were diagnosed with LCA6 caused by variation of the  RPGRIP1 gene. All five patients exhibited nystagmus and visual loss at or soon after birth, and case 2 and case 3 had similar morphological characteristics in the fundus photographs and OCT. Case 2 belonged to group 1, and case 3 belonged to group 2 for the ERGs. Furthermore, both case 1 and case 2 belonged to group 1 for the electrophysiological response, while they had a different fundus appearance and fundus autofluorescence, even for the OCT. According to the literature, the fundus appearance of patients with RPGRIP1 associated with LCA is highly variable, and the fundus photographs have mentioned disc pallor, attenuated vessels, the pigment epithelium mottling, or bone spicules from the mid-retina to the periphery [14][15][16][17][18], sometimes with a grossly normal appearance [3,16,18]. e funduscopy of our patients with LCA6 has been reported as mentioned above, but some different information which was symmetry atrophy of the retina with mottled pigmentary at the inferior-nasal side had been discovered in case 1. ere have been no reports regarding fundus photographs until now. Other researchers have only reported the high variability of fundus appearance, and there are no reports on RPGRIP1-related LCA6 patients' functions. us, we assessed the patient's local and overall visual function in detail, such as perimetry, mfERG, ffERG, and FVEP. We found that there was an inconsistency in the morphology and function among case 1, case 2, and case 3. erefore, different examination methods can reflect the patient's progress and the morphological and functional differences of different patients with the same disease. Due to their different principles and characteristics, various types of examinations had significance to our subsequent clinical diagnoses. e fundus manifestations and clinical symptoms of case 4 and case 5 were typical LCA6. e OCT indicated that the lamellar structure in the foveal area is retained to some extents from case 1 to case 3, with the exception of case 4. Blue arrows revealed that the ellipsoid zone was partially retained in case 1, a large area was preserved for case 2, and the ellipsoid zone was blurred  A recent study showed that all patients' phenotypes appeared to have a thinning of the photoreceptor outer nuclear layer (ONL) and that a retinal lamellar structure was maintained, whereas the ellipsoid zone was blurred or disappeared [18]. In addition, Suzuki et al. and Imani et al. described RPGRIP1-related LCA patients as having a loss of a normal foveal configuration, which is similar to case 4 [16,19]. e phenotypic variability was high. OCT can only reflect the macular area, and it cannot reflect the morphology or function in other areas of the retina. erefore, it is necessary to diagnose different types of hereditary retinopathy using different examination methods, which involves partial or overall phenotypes between the morphology and function. e RPGRIP1 gene is located on chromosome 14q11 and contains 24 exons. It encodes a protein which expresses in the cilium, connecting the outer and inner segments of human photoreceptor [20,21]. A previous study revealed that mice lacking RPGRIP1 showed high defects in disc formation photoreceptor outer segments and photoreceptor loss at 20 days and 5 months of age, respectively [5]. Moreover, a better preservation of photoreceptor function in the treated eyes after gene therapy on a murine model with RPGRIP1 variants has been observed by an ERG and histological examination [6]. Lhériteau et al. provided great promise for human treatment: gene addition therapy can restore the functional deficit of cone photoreceptors and prevent the retinal degeneration and vision loss in RPGRIP1deficient dogs, a model exhibiting a severe cone-rod dystrophy similar to that seen in humans [22]. Overall, RPGRIP1 is indispensable in the disc morphogenesis of photoreceptors. To date, in HGMD Professional, 87 missense/nonsense variations in RPGRIP1 have been reported. Among them, 27 (59%) of the 47 missense variations are located at the C2 domain and the following was RID, which occupied 6 (13%). is indicates that the C2 domain could likely be a hot spot for missense variants. In this study, we found that two novel missense variants (p.T652A and p.D862G) are associated with the C2 domain (Supplemental Figure 3). Moreover, the clinical symptoms of case 4 and case 5 were the most severe and may be related to the type of variation they carried, which was predicted to be loss-offunction variation. No genotype to phenotype correlation could be established in our cohort, but the disease-causing gene of RPGRIP1 in these five LCA6 subjects diagnosed was rigorously cosegregated with the genotype and phenotype from 352 unrelated families with inherited retinal degeneration and expands the variant spectrum of RPGRIP1.
erefore, future studies with a larger sample size of LCA patients with RPGRIP1 variants should be conducted using longitudinal clinical assessments and a genotype-phenotype correlation analysis.
In conclusion, we identified the heterogeneous retinal morphology changes and visual functions of RPGRIP1-LCA6 patients. e combination of perimetry, ffERG, FVEP, and other detection methods not only help in objectively assessing the fundus morphology and visual pathway of LCA6 patients but also will aid in the application of visual function evaluations after treatment in the future. We will conduct a large-scale screening LCA6 clinical cohort study among LCA patients in China to establish genotype to phenotype correlations to explore new treatments, such as gene therapy, as well as genetic counseling.

Data Availability
e data used to support the findings of this study are included within the article.

Disclosure
Yumei Mao and Yanling Long are the co-first authors.

Conflicts of Interest
e authors declare no conflicts of interest.