ERG indicates electroretinographic; HET, heterozygous; OCT, optical coherence tomographic; OD, right eye; OS, left eye; and MU, mutation.
HET indicates heterozygous; HOM, homozygous; MU, mutation; and WT, wild type.
OCT indicates optical coherence tomographic; OS, left eye.
ERG indicates electroretinographic; OD, right eye; OS, left eye; and MU, mutation.
Typical presentations of retinitis pigmentosa on fundus photographs can be seen for patients RH1-V:1, RH1-V:2, RH4-II:6, and RH17-IV:2. Macular degeneration can been in the fundus photographs of patients RH9-V:1, RH9-V:2, RH9-V3, and RH10-II:1.
HET indicates heterozygous; OCT, optical coherence tomographic; MRI, magnetic resonance imaging; and MU, mutation. The white arrowheads indicate bone spicular pigmentation. The yellow arrowheads indicate enlargement of the central canal from C3 to T8 vertebral levels.
eFigure 1. Pedigree structure of the other 9 families with negative results
eFigure 2. Splicing abnormality caused by the FLVCR1 mutation
eTable 1. Overall information of the 20 investigated Chinese families
eTable 2. Clinical features of attainable patients in 9 families with negative results
eTable 3. Variants excluded from the remaining 9 families
eTable 4. Overview of data production
Liu X, Xiao J, Huang H, Guan L, Zhao K, Xu Q, Zhang X, Pan X, Gu S, Chen Y, Zhang J, Shen Y, Jiang H, Gao X, Kang X, Sheng X, Chen X, Zhao C. Molecular Genetic Testing in Clinical Diagnostic Assessments That Demonstrate Correlations in Patients With Autosomal Recessive Inherited Retinal Dystrophy. JAMA Ophthalmol. 2015;133(4):427-436. doi:10.1001/jamaophthalmol.2014.5831
Inherited retinal dystrophies (IRDs) are a group of retinal degenerative diseases presenting genetic and clinical heterogeneities, which have challenged the genetic and clinical diagnoses of IRDs. Genetic evaluations of patients with IRD might result in better clinical assessments and better management of patients.
To determine the genetic lesions with phenotypic correlations in patients with diverse autosomal recessive IRD using next-generation sequencing.
Design, Setting, and Participants
A cohort of 20 Chinese families affected with autosomal recessive IRD were recruited (with data on their detailed family history and on their clinical condition). To identify disease-causing mutations in the patients, the targeted sequence capture of IRD-relevant genes using 2 in-house–designed microarrays, followed by next-generation sequencing, was performed. Bioinformatics annotation, intrafamilial cosegregation analyses, in silico analyses, and functional analyses were subsequently conducted for the variants identified by next-generation sequencing.
Main Outcomes and Measures
The results of detailed clinical evaluations, the identification of disease-causing mutations, and the clinical diagnosis.
Homozygous and biallelic variants were identified in 11 of the 20 families (55%) as very likely disease-causing mutations, including a total of 17 alleles, of which 12 are novel. The 17 alleles identified here include 3 missense, 6 nonsense, 4 frameshift, and 4 splice site mutations. In addition, we found biallelic RP1 mutations in a patient with cone-rod dystrophy, which was not previously correlated with RP1 mutations. Moreover, the identification of pathogenic mutations in 3 families helped to refine their clinical diagnoses.
Conclusions and Relevance
In this study, to our knowledge, many mutations identified in those known loci for autosomal recessive IRD are novel. Specific RP1 mutations may correlate with cone-rod dystrophy. Genetic evaluations with targeted next-generation sequencing might result in a better clinical diagnosis and a better clinical assessment and, therefore, should be recommended for such patients.
Inherited retinal dystrophies (IRDs) are a group of diverse retinal degenerative diseases with both genetic and clinical heterogeneities. Clinically, IRDs manifest as isolated retinal degeneration or as a systemic disease with retinal dystrophy. Isolated IRDs include retinitis pigmentosa (RP),1 cone-rod dystrophy (CRD),2 cone dystrophy,3 congenital stationary night blindness,4 Leber congenital amaurosis,5 Bietti crystalline dystrophy,6 Stargardt disease,7 Best vitelliform macular dystrophy,8 and various other comparatively rare retinal degenerations. Typical systemic diseases that may be accompanied with IRD may include Usher syndrome (OMIM 276900) or Bardet-Biedl syndrome (OMIM 209900), as well as others. Clinical diagnoses of IRDs can sometimes be challenged by phenotypic overlaps among distinct diseases and among certain conditions, such as a young child with a less severe clinical condition or a retinal disease that is possibly a part of a syndrome. In such situations, molecular genetic testing could be useful to address the clinical ambiguity in diagnosis.
Inherited retinal dystrophy can be inherited via all 3 methods of Mendelian inheritance (ie autosomal dominant, recessive, and X-linked patterns). Digenic, mitochondrial, and incomplete dominant forms have also been reported.1,9,10 To date, 261 loci (including 221 identified genes) have been implicated in the etiology of IRDs (see the Retinal Information Network [RetNet] at https://sph.uth.edu/retnet/), representing its great genetic heterogeneity. Traditional approaches to the detection of mutations have their limitations, resulting in low diagnostic rates. However, targeted next-generation sequencing (NGS) enables parallel sequencing of a panel of numerous candidate genes and has been proved to be an efficient tool for the molecular diagnosis of various IRDs11 and of autosomal recessive RP.12
To develop an effective genetic diagnostic tool for IRDs, we have previously used a targeted NGS approach, by which we identified disease-causing mutations in multiple types of IRDs.11,13,14 Herein, we have further applied this approach in the investigation of a cohort of 20 Chinese families with autosomal recessive IRD.
Our study conformed to the tenets of the Declaration of Helsinki and was prospectively reviewed and approved by the local institutional review boards. Written informed consent was obtained from all participants or their legal guardians. A cohort of 20 unrelated Chinese families, including 33 patients affected with IRD and 61 unaffected family members, were recruited from multiple hospitals in China (eTable 1 in the Supplement). Family histories and personal medical records were carefully checked and reviewed. All of the participants underwent detailed ophthalmic examinations at the beginning of the study and systemic examinations when necessary. The data obtained from these ophthalmic examinations included best-corrected visual acuity, results of a slitlamp examination, intraocular pressure, results of a funduscopic examination, visual field, and electroretinograms. Optical coherence tomography was performed for patients with macular degeneration. For patient RH13-II:1, a 3-T magnetic resonance scanner (Magnetom Trio; Siemens Medical Solutions) with a transmit-receive extremity coil was used for an examination of the spinal cord. In addition, 150 unrelated Chinese controls without IRD or other ocular diseases were also included. Samples of venous blood (5 mL) were obtained from each participant for genomic DNA isolation, which was performed using a QIAmp DNA Mini Blood Kit (Qiagen).
A targeted gene approach was completed using 2 previously described capture arrays (from Roche NimbleGen). Microarray 1 was designed to capture the targeted region of 179 IRD-related genes and 10 candidate genes.11 Microarray 2 was designed to capture the coding sequence region of 316 genes related to inherited ocular diseases.15 Sequence capture, enrichment, elution, and NGS were conducted as detailed previously.16 For bioinformatics analyses, the results of Sanger sequencing, in silico analyses, and the results of reverse transcription–polymerase chain reaction, see the eAppendix in the Supplement.
Here, we only focus on the results of 11 of the 20 families investigated because putative disease-causing mutations were identified in the 11 families. Detailed clinical data on the 11 families are summarized in Table 1 and Table 2, whereas the clinical details of family pedigrees and the genetic findings of the other 9 families are presented in eFigure 1 and eTables 2 and 3 in the Supplement. One or 2 family members from each family were selected for NGS, and the results of NGS are detailed in eTable 4 in the Supplement. In brief, a total of 26 650 variants were initially detected by targeted NGS in the 11 families. Of all tested samples, the mean call rate of the targeted region was about 99.8%, and the mean depth was about 91.3-fold. A total of 17 homozygous or compound heterozygous variants in the 11 families passed the filtration process, of which 12 were novel and 5 were previously reported mutations (Table 3). Those novel variants were absent in 150 unrelated Chinese controls. The potential pathogenicity of those novel putative mutations was evaluated by multiple in silico programs and is summarized in Table 3.
Novel biallelic mutations in RP1 (p.[E474Gfs*11]; [K1939*]) were identified in patient RH15-II:1 with CRD (Figure 1). This patient was reported to have central visual defects since the age of 5 years, and her best-corrected visual acuity was 20/100 for the right eye and 20/400 for the left eye at her last visit to our hospital. Macular degeneration was revealed in fundus photographs and optical coherence tomographic images (Figure 1B and C). Reduced photopic and scotopic electroretinographic responses were observed (Figure 1D). Visual field constriction and central scotoma were also indicated (Figure 1E).
CYP4V2 mutation c.802-8_810del17insGC, a frequent mutation in East Asian populations,21- 23 was identified in 3 families (RH6, RH12, and RH18) (Figure 2A). This mutation, located in the flanking intronic region (Figure 2B), has been proven to cause the skip of whole exon 7 by reverse transcription–polymerase chain reaction,24 thus generating a truncated protein (p.I260_N339del) lacking 80 in-frame amino acids. This mutation was found to be homozygous in the patients from families RH6 and RH12 with Bietti crystalline dystrophy, whereas biallelic CYP4V2 mutations (p.[I260_N339del]; [H331P]) were found in patients from family RH18 with typical RP (Figure 2A and B; Figure 3A). Conservational analysis also proved the high level of conservation of residue H331 in multiple orthologous protein sequences (Figure 2C). Clinical evaluations were presented in Figure 3A and B and Tables 1 and 2, and waxy optic discs and attenuated vessels are present in all patients. Crystal deposits are present in patients RH6-II:1, RH6-II:2, and RH12-II:3, whereas bone spicule–like pigments are revealed in patients RH12-II:3, RH18-II:1, RH18-II:2, and RH18-II:3. Macular atrophy was detected on the optical coherence tomographic images of patient RH6-II:1. An interesting finding is that intensive bone spicule–like pigments were observed in the fundus photographs of patient RH12-II:3, whereas crystal deposits were only found in the macular region (Figure 3A).
A previously reported homozygous missense variant in BBS2 (p.R413* [CM033336]) was identified in family RH1 with Bardet-Biedl syndrome (Figure 4A). The 2 siblings, RH1-V:1 and RH1-V:2, were still in their teens when they first were referred to our clinic for having poor night vision and a constricted visual field. Results of an ophthalmic examination revealed typical RP presentations (Figure 5). Other than ophthalmic abnormities, results of systemic examinations revealed polydactylism and mild cognitive impairment in both patients. The affected sister showed no signs of menophania at her visit when she was 18 years of age. Each member of this family thus received a diagnosis of Bardet-Biedl syndrome based on genetic and clinical findings.
A novel homozygous frameshift mutation in PROM1 (p.K549Qfs*2) was identified in patients from family RH9 (Figure 4A). PROM1 mutations have been reported to be the cause of recessive RP with macular degeneration, dominant Stargardt disease–like macular dystrophy, dominant cone dystrophy, and dominant CRD (RetNet). All 3 patients from family RH9 reported having night blindness since early childhood, and at their last visit, each patient had central vision that was severely impaired owing to severe macular degeneration (Figure 5; Tables 1 and 2). Presentations of RP and macular degeneration were indicated in the fundus photographs of patients from this family. Therefore, based on the genetic and clinical findings, we finalized the clinical diagnosis to autosomal recessive RP with macular degeneration for this family.
Novel biallelic ALMS1 mutations p.[S1218*]; [Y1533*] were identified as disease causing for family RH10 (Figure 4A; Tables 1-3). Patient RH10-II:1 was reported to have nystagmus and photophobia since 5 months of age. His visual acuity began to decrease rapidly early in the second decade of his life. At his last visit to our clinic, he was 18 years of age when his best-corrected visual acuity was light perception for both eyes. Fundus photographs and electroretinographic responses indicated a typical presentation of CRD in this patient (Figure 5). Because ALMS1 mutations have been reported to cause Alström syndrome, his medical records were further examined, and more biochemical tests performed. Detailed systematic examinations revealed sensorineural hearing loss, early-onset type 2 diabetes mellitus, obesity, dilated cardiomyopathy, and hepatic dysfunction in this patient, which fully met the diagnostic criteria for Alström syndrome. Therefore, the genetic findings for this patient helped to define the clinical diagnosis of Alström syndrome in this case.
Patient RH13-II:1 had poor night vision since early childhood, followed by a rapid decrease in her visual field and central vision (Figure 6A; Tables 1 and 2). Ophthalmic evaluations demonstrated typical RP phenotypes with macular edema in her right eye (Figure 6B and C). Complex neural phenotypes were also noted in patient RH13-II:1, including mild ataxia since childhood, attenuation of deep tendon reflexes, and superficial sensations. Magnetic resonance imaging revealed a mild enlargement of the central canal from C3 to T8 vertebral levels. Herniated disks were indicated in C4/5, C5/6, and C6/7, and magnetic resonance imaging at C4/5 revealed the centrally herniated disk material and the narrowing of the spinal canal (Figure 6E-G).
By use of targeted NGS, we identified novel biallelic mutations in the FLVCR1 gene in patient RH13-II:1, including the paternal inherited splice site mutation c.883+6T>C and the maternal inherited missense variation p.G384R. The missense variation was predicted to be deleterious by all in silico programs, and the residue G384 was absolutely conserved through evolution (Figure 1A and D). Reverse transcription–polymerase chain reaction was then performed to determine the effect of the paternal inherited allele. Reverse transcription–polymerase chain reaction products were separated by agarose gel electrophoresis, and 4 bands were observed in patient RH13-II:1 and her unaffected father RH13-I:1, who carried the same allele (eFigure 2A in the Supplement). Sequencing of the product revealed that this splice site mutation would cause aberrant splicing of the FLVCR1 gene by generating a mutant complementary DNA fragment inserted with a 127–base pair (bp) fragment from intron 2 beginning at c.883+524 (eFigure 2B in the Supplement). FLVCR1 mutations have been implicated in the disease etiology of autosomal recessive posterior column ataxia with RP (PCARP).25- 27 Therefore, this patient received a diagnosis of PCARP.
The novel homozygous missense mutation in CRB1 (p.C606R) was identified in patient RH17-IV:3 with RP, whose parents were first-degree cousins. Typical RP presentations on fundus photographs, including waxy pallor of optic discs, attenuated vessels, and bone spicular pigmentations, were demonstrated by this patient (Figure 5). Significantly reduced scotopic and photopic electroretinographic responses are shown for patient RH17-IV:3 (Figure 4C). According to RetNet, CRB1 mutations have been implicated in a wide panel of IRDs, including recessive RP, recessive Leber congenital amaurosis, and dominant pigmented paravenous chorioretinal atrophy. The affected residue C606 was conserved among all mammal species and is located in the first laminin G–like domain of the protein Crumbs homologue 1 encoded by the CRB1 gene (Figure 4B). The nature of p.C606R, a missense mutation, may explain why it correlates with recessive RP but not with more severe diseases such as recessive Leber congenital amaurosis or dominant retinal dystrophy.
Biallelic mutations in ABCA4 (p.[Y808*]; [V521Sfs*46]) were found to cause Stargardt disease in patient RH4-II:6 (Figure 4A; Figure 5; Tables 1-3), and novel biallelic mutations in CNGA1 (p.[G133Vfs*28]; [R560*]) were revealed in patient RH20-II:1 with typical RP (Figure 4A). ABCA4 mutations have been found to cause recessive Stargardt disease, recessive macular dystrophy, recessive RP, recessive fundus flavimaculatus, and recessive CRD, whereas CNGA1 mutations were only implicated in the disease etiology of RP (RetNet). Therefore, the correlations between the affected genes and the phenotypes observed in the 2 families were previously established.
We have previously evaluated the efficiency of targeted NGS in families with IRD, which is mainly in autosomal dominant trait or sporadic cases.11 In the present study, we focused on the mutation analyses for a cohort of 20 Chinese families with autosomal recessive IRD. A detection rate of 55% (11 of 20 families) is achieved in this cohort, similar to that in our previous study (56%)11 and a bit higher than that in other Chinese cohorts with autosomal recessive IRD.12,28,29 Seventeen disease-causing mutations have been identified in the 11 families investigated, among which 12 are novel to our knowledge. Thus, our study extends the spectrum of IRD disease-causing mutations.
Mutations in RP1 have been reported to cause RP, accounting for 5.5% of the autosomal dominant form and 1% of the recessive form.30 The RP1 gene encodes a protein of 2156 amino acids and is located in the connecting cilia of both rod and cone photoreceptors.31 The RP1 protein contains 2 doublecortin (DCX) domains (residues 36 to 118 [DCX1] and 154 to 233 [DCX2]), via which the RP1 protein interacts with microtubules. The RP1 protein is thus the first identified photoreceptor-specific and microtubule-associated ciliary protein, which functions in the organization of the photoreceptor outer segments to ensure the exact orientation and higher-order stacking of outer segment disks along the photoreceptor axoneme.32 Close interactions between RP1 and other ciliary proteins, including the Rp1-like protein (RP1L1) and the male germ cell–associated kinase, have also been identified.32,33RP1L1 mutations have been implicated in occult macular dystrophy34 and RP,35 suggesting a potential diverse role of RP1 in IRDs. By far, RP1 mutations have only been reported to cause RP, whereas, in the present study, we have identified biallelic mutations in RP1 (p.E474Gfs*11 and p.K1939*) in patient RH15-II:1 with CRD. Similar to our findings, most identified RP1 mutations are protein-truncating mutations resulting in premature termination codons, which would lead to RP1 defects via the generation of truncated proteins or nonsense-mediated messenger RNA decay.36,37 Thus, we have identified RP1 mutations in patients with CRD, suggesting a potential relationship between RP1 defects and the pathogenesis of CRD.
Bietti crystalline dystrophy, also referred to as Bietti crystalline corneoretinal dystrophy, is inherited in an autosomal recessive fashion with only 1 disease-causing gene, CYP4V2.38 Patients with Bietti crystalline dystrophy often present with corneal crystals, yellow deposits in the retina, and progressive retinal and choroidal atrophy.39 Other than Bietti crystalline dystrophy, biallelic mutations in CYP4V2 have been reported in Chinese families with recessive RP,12,23 one of which shows complicated phenotypes, including RP, thin corneas, congenital cataracts, and high myopia.23 In our study, we have identified biallelic CYP4V2 mutations c.802-8_810del7insGC (p.I260_N339del) and c.992A>C (p.H331P) in a Chinese family with RP. These mutations are frequently seen in East Asian populations. The ages at onset of disease for all 3 patients in this family are similar to those for patients in the previously reported families.12,23 No other ophthalmic abnormalities were observed. Cytochrome P450 4V2, the protein encoded by the CYP4V2 gene containing 525 amino acids, plays a crucial role in the metabolism of fatty acids and steroids in the eye,40 particularly in the hydroxylation of the omega-3 polyunsaturated fatty acids, including docosahexaenoic acid and eicosapentaenoic acid.41 Polyunsaturated fatty acids function in the renewal of disk membranes in the outer segments of photoreceptor cells and demonstrate a much higher level of expression in the retina when compared with other tissues.42 Thus, CYP4V2 defects would probably lead to disease via the dysfunction or deficiency of polyunsaturated fatty acids. Dietary supplementation of polyunsaturated fatty acids in such patients would possibly help to slow down the progression of disease and might be recommended by clinicians.
The human feline leukemia virus subgroup C receptor 1 (hFLVCR) protein, encoded by the FLVCR1 gene, is a heme exporter protein crucial for maintaining the intracellular concentration of heme. As a cell surface receptor, hFLVCR mainly functions in heme export and erythroid maturation. Inhibition of hFLVCR will cause apoptosis of erythroid cells. Six mutations in the FLVCR1 gene have been previously reported to cause PCARP.25- 27 In our study, compound heterozygous mutations in FLVCR1, c.[883+6T>C]; [1150G>C] (p.[?]; [Gly384Arg]), are found in a Chinese patient with PCARP. Similar to previous findings, these 2 mutations were also located in the transmembrane domains, indicating a pathogenesis similar to that of the previously investigated mutations.43 Thus, we hypothesize that mutant hFLVCR with p.Gly384Arg would also misfold in the endoplasmic reticulum, partially degrade in lysosomes, lose heme export activity, cause intracellular accumulation, and lead to apoptosis.43 Furthermore, the splice site mutation, c.883+6T>C, is predicted to generate a splice donor site and a splice receptor site within the intronic sequence between exons 2 and 3, thus leading to the insertion of a 127-bp fragment. This insertion would probably generate an altered protein with an irregular function or cause messenger RNA decay. Of note, further investigation of this inserted fragment has revealed its existence as the second exon of the coding sequence of another FLVCR1 transcript (ENST00000579295); however, no protein product has been annotated for this transcript. Because the specific mechanism underlying the association between FLVCR1 gene deficits and symptoms of PCARP is still unclear, we hypothesize a potential linkage between hFLVCR protein dysfunction and neurological problems.
Therefore, our study demonstrates that the targeted NGS approach might help to generate a molecular diagnosis for autosomal recessive IRDs. However, this approach also has limitations. A mean mismatch rate of 0.234% was reached in our study, indicating the potential existence of false-positive or false-negative variants. In addition, deep intronic variations and copy number variations could not be detected by this approach. No disease-causing mutations were identified in the remaining 9 families. These 9 families may carry mutations in unknown loci, deep intronic mutations, or copy number variations, all of which are not detectable by our targeted NGS approach.
In conclusion, by means of targeted NGS, we have revealed 15 disease-causing mutations, including 12 novel mutations found in 11 of 20 Chinese families with autosomal recessive IRD. Our finding reaches a detection rate of 55% in the investigated cohort, which demonstrates the efficiency of targeted NGS in analyzing the etiology for autosomal recessive IRDs. In addition, based on all our findings, we believe that the genetic evaluations would help with clinical assessments.
Submitted for Publication: August 20, 2014; final revision received November 26, 2014; accepted November 30, 2014.
Corresponding Author: Chen Zhao, MD, PhD, Department of Ophthalmology, The First Affiliated Hospital of Nanjing Medical University, State Key Laboratory of Reproductive Medicine, 300 Guangzhou Rd, Nanjing, Jiangsu 210029, China (firstname.lastname@example.org).
Published Online: January 22, 2015. doi:10.1001/jamaophthalmol.2014.5831.
Author Contributions: Dr C. Zhao had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Liu, Xiao, Huang, Guan, and X. Chen contributed equally to this article.
Study concept and design: K. Zhao, X. Chen, C. Zhao.
Acquisition, analysis, or interpretation of data: Liu, Xiao, Huang, Guan, Xu, X. Zhang, Pan, Gu, Y. Chen, J. Zhang, Shen, Jiang, Gao, Kang, Sheng.
Drafting of the manuscript: Liu, Xiao, Huang, Guan, Gu, Y. Chen, J. Zhang, Shen, Jiang, X. Chen.
Critical revision of the manuscript for important intellectual content: Liu, K. Zhao, Xu, X. Zhang, Pan, Gao, Kang, Sheng, X. Chen, C. Zhao.
Statistical analysis: Liu, Gu.
Obtained funding: C. Zhao.
Administrative, technical, or material support: All authors.
Study supervision: C. Zhao.
Conflict of Interest Disclosures: None reported.
Funding/Support: This work was supported by the National Key Basic Research Program of China (program 973, grant 2013CB967500), the National Natural Science Foundation of China (grants 81222009, 81170856, 81260154, and 81170867), the Thousand Youth Talents Program of China (to Dr C. Zhao), the Jiangsu Outstanding Young Investigator Program (grant BK2012046), the Jiangsu Province’s Key Provincial Talents Program (grant RC201149), the Fundamental Research Funds of the State Key Laboratory of Ophthalmology (to Dr C. Zhao), the Jiangsu Province’s Scientific Research Innovation Program for Postgraduates (grant CXZZ13_0590), and a project funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions.
Role of the Funder/Sponsor: The funders/sponsors had no role in the design and conduct of the study; collection, management, analysis, or interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.