Mutation spectrum in a cohort with familial exudative vitreoretinopathy

Abstract Purpose To expand the mutation spectrum of patients with familial exudative vitreoretinopathy (FEVR) disease. Participants 74 probands (53 families and 21 sporadic probands) with familial exudative vitreoretinopathy (FEVR) disease and their available family members (n = 188) were recruited for sequencing. Methods Panel‐based targeted screening was performed on all subjects. Before sanger sequencing, variants of LRP5, NDP, FZD4, TSPAN12, ZNF408, KIF11, RCBTB1, JAG1, and CTNNA1 genes were verified by a series of bioinformatics tools and genotype–phenotype co‐segregation analysis. Results 40.54% (30/74) of the probands were sighted to possess at least one etiological mutation of the nine FEVR‐causative genes. The etiological mutation detection rate was 37.74% (20/53) in family‐attainable probands while 47.62% (10/21) in sporadic cases. The diagnosis rate of patients in the early‐onset subgroup (≤5 years old, 45.4%) is higher than that of the children or adolescence‐onset subgroup (6–16 years old, 42.1%) and the late‐onset subgroup (≥17 years old, 39.4%). A total of 36 etiological mutations were identified in this study, comprising 26 novel mutations and 10 reported mutations. LRP5 was the most prevalent mutant gene among the 36 mutation types with a percentage of 41.67% (15/36). Followed by FZD4 (10/36, 27.78%), TSPAN12 (5/36, 13.89%), NDP (4/36, 11.11%), KIF11 (1/36, 2.78%), and RCBTB1 (1/36, 2.78%). Among these mutations, 63.89% (23/36) were missense mutations, 25.00% (9/36) were frameshift mutations, 5.56% (2/36) were splicing mutations, 5.56% (2/36) were nonsense mutations. Moreover, the clinical pathogenicity of these variants was defined according to American College of Medical Genetics (ACMG) and genomics guidelines: 41.67% (15/36) were likely pathogenic variants, 27.78% (10/36) pathogenic variants, 30.55% (11/36) variants of uncertain significance. No etiological mutations discovered in the ZNF408, JAG1, and CTNNA1 genes in this FEVR cohort. Conclusions We systematically screened nine FEVR disease‐associated genes in a cohort of 74 Chinese probands with FEVR disease. With a detection rate of 40.54%, 36 etiological mutations of six genes were authenticated in 30 probands, including 26 novel mutations and 10 reported mutations. The most prevalent mutated gene is LRP5, followed by FZD4, TSPAN12, NDP, KIF11, and RCBTB1. In total, a de novo mutation was confirmed. Our study significantly clarified the mutation spectrum of variants bounded up to FEVR disease.


| INTRODUCTION
Familial exudative vitreoretinopathy (FEVR, OMIM 133780) is a rare genetic disorder which is characterized by deviant development of peripheral retinal vessels (Criswick & Schepens, 1969;Poulter et al., 2016;van Nouhuys, 1989). Criswick and Schepens firstly described this disease in 1969 (Criswick & Schepens, 1969). The clinical manifestations are widely variable in diverse FEVR patients, ranging from asymptomatic peripheral vascular abnormalities to congenital blindness (van Nouhuys, 1989). The sight-threatening features of the FEVR phenotype are considered secondary to retinal avascularity and develop because of the resulting retinal ischemia; they include the development of hyperpermeable blood vessels, neovascularization, vitreoretinal traction, retinal folds, and retinal detachments (Poulter et al., 2016). Based on the diverse clinical patterns (including the presence avascular zone in the extreme periphery, retinal detachment, neovascularization, macular ectopia, falciform retinal fold, vitreoretinal adhesion, and exudative retinal detachment) in FEVR patients, a grading system was proposed to divide FEVR disorder into five types (Miyakubo et al., 1984). FEVR can be inherited in patterns of autosomal dominant pattern, autosomal recessive, and X-linkage inheritance, and the most prevalent pattern of inheritance is autosomal dominant (Chen et al., 1993;Laqua, 1980;Rao et al., 2017).
In this study, we systematically screened nine reported FEVR disease-associated genes (LRP5, FZD4, TSPAN12, NDP, KIF11, ZNF408, RCBTB1, JAG1, and CTNNA1) in a cohort of 74 Chinese probands and their available family members. The diagnosis of the FEVR patients was performed by professional ophthalmologists. We amplified the mutation spectrum of patients with FEVR disease which can not only benefit clinical practice but also assist in designing the ophthalmic panel.

| Subjects collection and clinical assessment
This study was approved by the Ethics Committee of the Eye & ENT Hospital of Fudan University and complies with the tenets of the Declaration of Helsinki for medical research related to human subjects. All participants had signed the written informed consent. For minors, obtain the consent from the legal guardian. 74 probands and their available family members (total participants: 188) who came to Eye & ENT Hospital of Fudan University in 2017-2020 were recruited for this study. Encompassing ophthalmologic examination was performed on all subjects, including the best corrected visual acuity (BCVA), slit-lamp biomicroscopy, color vision testing, intraocular pressure, Humphrey perimetry, electroretinography, fundus autofluorescence, and spectral domain optical coherence tomography (ST-OCT). The final clinical diagnosis of FEVR patients was appraised by professional ophthalmologists. All these FEVR probands are divided into three groups according to the age of onset, including the earlyonset subgroup (≤5 years, n = 22), the children or adolescence subgroup (6-16 years, n = 19), and the late-onset subgroup (≥17 years, n = 33). The details of this FEVR cohort are depicted in Table 1.

| DNA sample collection
Peripheral blood samples were collected from all subjects. We used the Flexigene DNA Kit to exact genomic DNA. Collected DNA samples were stored at −20°C.

| Targeted exome sequencing
Panel-based next-generation sequencing was accomplished on all subjects recruited in this study. We designed a high-throughput chip that contains 792 eye-diseasesrelated genes (Supplementary Table S1) to actualize precise targeted sequencing. The Agilent SureSelect Target Enrichment Kit (Agilent Technologies, Inc., USA) was utilized to conduct DNA libraries and then the reads were sequenced on MGISEQ-2000 platform (DNBSEQ-G400) (Gao et al., 2019;Li et al., 2021). The size of the generated reads during paired-end sequencing was 100 bp and the average depth was 704X while the median depth was 677X.

| Statistical analysis
We calculated the detection yield and clarified the mutation spectrum in both the family-attainable cohort and the sporadic cohort. The pattern of inheritance in families was investigated and described in the "Results" section. We also depicted the spectrum in subjects with novel variants and reported variants. Novel mutation was defined if it had not been reported in previous literature. De novo mutation was ascertained if it did not appear in the biological parents of the mutation-carried patients. The paternity would be checked when we reported a de novo variant.
Of all the 24 mutations detected in the families, 12.50% (3/24) were inherited in the patterns of X-linkage recessive inheritance. Including three missense mutations (c.338G>A, c.137A>G, c.343C>G) of NDP gene. One de novo mutation (LRP5, c.4724C>G, p.P1575R) was uncovered in family 4 (F4). The remaining 20 mutations were inherited in autosomal dominant pattern. Among them, 10.0% (2/20) were considered as pathogenic mutations according to the methods we described in Materials and Methods, 40.0% (8/20) were determined to be likely pathogenic mutations and 50.0% (10/20) were thought to be variants of uncertain significance. No variants were sighted to be inherited in autosomal recessive pattern.
One novel missense mutation (c.1238T>G, p.I413S) of RCBTB1 gene was found in two members of family 8 (F8), the proband and his affected biological father. This mutation is located in a highly evolutionarily conserved region (Figure 3a) and altered the corresponding amino acid from isoleucine to serine. The 3D structural model of these amino changes is portrayed in Figure 3b. Using PyMOL (https://pymol.org/2/) to visualize the structure of peptide chain, and the results show that the αhelix of mutant protein is missing (Janson & Paiardini, 2020). No variants of the other eight genes were checked out in the two FEVR subjects. This variant was predicted to be likely pathogenic in accordance with ACMG and genomics guideline, hence we considered it as a candidate causative factor for the FEVR cases in this family.
We depict the comprehensive mutation spectrum of nine FEVR-causative genes and the detection yield is 40.54% (30/74) in FEVR cases. We uncover 26 novel variants of the six genes in total and 10 reported mutations. In the light of ACMG and genomics guidelines, 41.67% (15/36) of these variants were considered as likely pathogenic variants, 27.78% (10/36) were pathogenic variants, 30.55% (11/36) variants were defined as variants of uncertain significance. In correspondence with the previous reports, LRP5 accounts for the largest percentage of all the mutant genes in our study (Li et al., 2018;Tang et al., 2017). We divide the cohort into two groups: the probands with available family members and isolated cases. Different detection rates were revealed in these two groups: 37.74% (20/53) in family-attainable probands while 47.62% (10/21) in sporadic cases. 24 variants were detected in the familyattainable probands. Among them, 12.50% (3/24) inherited in the patterns of X-linkage recessive inheritance. One de novo mutation (LRP5, c.4724C>G, p.P1575R) was uncovered in family 4 (F4). The paternity in this family was checked seriously. The remaining 20 (83.33%, 20/24) variants were inherited in autosomal dominant pattern and no variants were sighted to be inherited in autosomal recessive pattern. Comparing the detection rate of FEVR probands in the three age groups, there was no significant difference between the early-onset subgroup, the child or adolescent subgroup, and the late-onset subgroup, indicating that genetic factors are the main cause of FEVR patients at all stages. A series of 389 consecutive FEVR patients from 389 families were sequenced for FZD4, LRP5, NDP, TSPAN12, ZNF408, and KIF11 genes. A total of 101 potentially pathogenic variants (PPV) and 49 variants of uncertain significance (VUS) were identified. Of them, 110 probands carried PPV (28.3%), and 51 probands carried VUS (13.1%). PPV in FZD4, LRP5, TSPAN12, NDP, ZNF408, and KIF11 accounted for 8.48%, 9.00%, 5.91%, 4.63%, 0.77%, and 0.77%, respectively (Li et al., 2018). Another study recruited 100 probands and their families for genetic screening of LRP5, NDP and TSPAN12 and identified 23 pathogenic mutations in 23 unrelated probands (10/23 in LRP5, 8/23 in TSPAN12, and 5/23 in NDP). The overall detection rate of the three known genes is 23%. The mutations of LRP5 and TSPAN12 are more frequent, accounting for 10% and 8%, respectively (Tang et al., 2017). The overall detection rate of the nine known genes in our study is 40.54% (30/74), and the high-frequency detection genes are LRP5 (41.67%), FZD4 (27.78%), TSPAN12 (13.89%), NDP (11.11%), KIF11 (2.78%), and RCBTB1 (2.78%).
Vitreoretinopathy is a group of genetic and clinically heterogeneous disorders in retinal vascular development. Recently, novel etiological genes have been discussed and reported. Two heterozygous frameshift mutations in the RCBTB1 gene were identified in three Taiwanese cases, and it was also proved that the RCBTB1 gene has a haploinsufficient mechanism, involved in the Norrin/FZD4 signaling pathway, and the transgenic fli1:EGFP zebrafish with rcbtb1 knockdown exhibited abnormalies in intraocular blood vessels (Wu et al., 2016). However, another refutation study identified biallelic mutations in RCBTB1 in an isolated patient with retinitis pigmentosa (RP, MIM 268000), and heterozygous truncated variants were distributed in different eye disease phenotypes but were not enriched in FEVR, emphasizing that it was not related to specific eye disease phenotypes . In F I G U R E 3 (a) multiple sequence alignment of different species of the mutation (the red arrow represents mutation sites). (b) 3D structural model of the wild type (WT) and mutant residues (RCBTB1 c.1238T>G). The red arrow represents the αhelix change of the peptide chain structure. our study, a missense mutation in the RCBTB1 gene was identified in the proband and the affected father of family 8 (F8), and multiple public databases and function prediction software were used for deleterious assessment. No other known mutations in FEVR-related genes have been detected, and the putative genetic cause of vitreoretinopathy phenotype is caused by the RCBTB1 gene. FEVR is a genetic and clinically heterogeneous disease. A reasonable explanation is that the RCBTB1 gene may cause the RP phenotype in the recessive mode, and the phenotype of vitreoretinopathy in dominant mode, sometimes as a syndrome or asymptomatic phenotype. Furthermore, the latest study elucidated that notch ligand JAG1 might be a novel candidate gene for FEVR (Zhang et al., 2020). This gene was also included in our panel and we screened this gene in all the subjects. However, no reliable pathogenic variants were founded in patients with FEVR disease. Besides, CNVkit was used to trace DNA copy number mutation in the whole cohort but no reliable result was discovered. 44 probands were negative for our 792 ophthalmic-disease-related genes-based panel sequencing. This phenomenon indicated that novel genes may be responsible for FEVR disease and the pathogenicity of JAG1 gene to FEVR disease need to be further investigated. A study of CTNNA1 identified three heterozygous mutations, demonstrating that FEVR-related mutations lead to excessive activation of Norrin/β-catenin signaling. After evaluating CTNNA1 mutations in the FEVR population, no potential pathogenic mutations have been found in our FEVR cohort (Zhu et al., 2021). Another CTNNB1 was reported to cause FEVR, but it was not included in our detection panel list (Dixon et al., 2016;Rossetti et al., 2021). In the future, we will further improve and update the range of genes covered by our panel.
In the study of 74 subjects from 17 separate families (including 17 patients and 57 family members), 43% of FEVR patients had detectable disease-causative mutations. However, only 8% of cohort patients reported a positive family history of FEVR in first-degree relatives. However, among asymptomatic family members, 58% of clinical or angiographic results are consistent with stage 1 or 2 FEVR, and 21% of clinical or angiographic results are consistent with stage 3, 4, or 5 FEVR (Kashani et al., 2014). Although the genetic pattern of FEVR is known, previous studies have not attempted to systematically recruit asymptomatic family members of FEVR patients for clinical and angiographic examinations and may underestimate the occurrence of the disease (albeit asymptomatic) (Pendergast & Trese, 1998;Ranchod et al., 2011). Obviously, clinical examination alone is not enough to identify the microvascular changes in early FEVR, and it is difficult to identify the subtle peripheral findings without angiography or insufficient angiography. The clinical significance of identifying patients with early and asymptomatic FEVR is not trivial. The identification of asymptomatic family members in children of childbearing age is important for potential genetic testing, newborn screening for future children, or both.
In conclusion, we significantly expand the mutation spectrum of patients with FEVR disease. Our study might cover the most FEVR-associated genes. Our comprehensive landscape can assist in understanding the genetic mechanism of FEVR and benefit clinical practice. With a detection yield of 40.54% (30/74), a total of 36 variants were identified after rigorous filtering, including one de novo mutation. We investigated the different mutation spectrum in the family-attainable cohort and the sporadic cases. Also, we ascertained the different patterns of inheritance of FEVR patients, including autosomal dominant pattern and X-linkage recessive inheritance. But no DNA copy number mutation were identified in the whole FEVR cohort. Besides, no reliable variants of JAG1 and CTNNA1 genes were detected in all the FEVR patients. Based on these results, further research about the genotype and phenotype relationship is needed to be invested in the future therefore we can better understand the mechanism of FEVR disease.