Identification of a Novel RPGR Exon ORF15 Mutation in a Family With X-linked Retinitis Pigmentosa

Objective  To investigate the phenotypic and genotypic characteristics of a novel mutation associated with X-linked retinitis pigmentosa (XLRP).

Methods  Six individuals in a family with XLRP were recruited, and clinical examinations were performed. All of the members were genotyped with microsatellite markers at loci that were considered to be associated with XLRP. The retinitis pigmentosa GTPase regulator gene (RPGR) was comprehensively screened using direct polymerase chain reaction sequencing.

Results  Genotyping analysis showed that the affected individuals in the family shared a common haplotype with selected markers. The patients demonstrated severe retinal degenerative phenotypes consistent with XLRP. Mutational screening of RPGR demonstrated a novel mutation, g.ORF15 + 1232_1233delGG.

Conclusions  We identified a novel mutation in the 3′ end of a highly repetitive region of exon open reading frame 15 (ORF15) and documented the detailed phenotypes of the patients with XLRP with the mutation. The clinical phenotype was consistent with XLRP, supporting the observation that the mutations in the 3′ end of the ORF15 coding sequence give rise to XLRP.

Clinical Relevance  The mutation in the 3′ end of the ORF15 coding sequence can lead to a spectrum of phenotypes, and the cone-predominant phenotype-related mutations can be located irregularly in exon ORF15.

Retinitis pigmentosa (RP) is a clinically and genetically heterogeneous disease with progressive degeneration of retinal photoreceptors characterized by night blindness, progressive loss of peripheral vision, and typically the appearance of bone spicules in the retina. X-linked RP (XLRP) is the most severe among different heritable forms of RP. In affected males, rod and cone dysfunctions begin in early childhood and functions deteriorate rapidly. Female carriers display a wide spectrum of clinical features, ranging from normal to severe manifestations.

To date, more than 32 genes responsible for RP have been cloned. Two of the most prevalent genes, the retinitis pigmentosa GTPase regulator gene (RPGR) and the retinitis pigment 2, X-linked gene (RP2), are known to be defective in 60% to 90% and 10% to 20% of families with XLRP, respectively.^1-4 The newly identified exon open reading frame 15 (ORF15), spanning a 1706–base pair (bp) coding sequence, is a mutation hot spot for XLRP.⁵ More recently, ORF15 mutations have been identified in X-linked cone-rod dystrophy (XLCORD).^6-9 A study¹⁰ involving comprehensive molecular screening of a large number of families with XLRP and several families with XLCORD identified a number of mutations. The investigators assumed that mutations close to downstream of ORF15 implicate the early preferential loss of cone function with slight-to-moderate loss of rod function and downstream mutations are responsible for XLCORD. A recent study⁹ reported 2 nonsense mutations located at the 3′ end of the highly repetitive region (codons 143-468) of ORF15 (codons 365 and 392), supporting the idea that mutations toward the 3′ end of the highly repetitive region may cause cone-predominant dysfunction. Thus far, RPGR exon ORF15 is the only known gene responsible for XLCORD. A number of mutations in the 3′ end of the ORF15 coding sequence have been identified in families with XLRP, but few detailed phenotypic descriptions have been documented.¹¹ Such data are important both for studying the genotype-phenotype correlation and for elucidating the distribution and contribution of the ORF15 mutations.

The aim of this study was to identify the mutation in a specific family with XLRP and to characterize the phenotypic changes in patients with the mutation. Through genotyping and mutational analysis, a novel mutation located at the 3′ end of the highly repetitive region of exon ORF15 was identified and the genotype-phenotype correlation was determined.

This study followed the tenets of the Declaration of Helsinki. The protocol of this study was approved by the ethics committee of Miyazaki Medical College. Informed consent was obtained from all of the family members participating in this study. The study consisted of 6 members who were recruited for DNA testing (Figure 1). Blood samples were collected and genomic DNA was extracted by standard protocols (DNA Extractor WB Kit; Wako Pure Chemical Industries, Ltd, Osaka, Japan). Clinical examinations included routine ophthalmic examinations, Goldmann perimetry, electroretinography (ERG), and color fundus photography. The diagnosis of RP was confirmed by ophthalmologists (Z.-B.J. and N.N.). There was no history of other ocular or systemic abnormalities in the family.

Genotyping using microsatellite markers (DXS1068, DXS8025, DXS6810, DXS8054, and G10578) involved polymerase chain reaction amplification of the microsatellite region following standard methods and measurement of the size of the amplified fragment. The oligonucleotide primer sequences and the order of the markers were taken from NCBI and ENSEMBL databases. The genotypes were obtained by silver stain and manual inspection. The pedigree and haplotypes were constructed by Cyrillic software version 2.1 (Cyrillic Software, Oxfordshire, England).

The coding fragments and intron-exon boundaries of RPGR, including exons 1 to 19, exon 15a, and ORF15, were amplified by polymerase chain reactions and sequenced according to the protocols described previously.^12-15 Nucleotide positions were based on GenBank sequence AF286472. The DNA for 118 normal subjects was screened as described previously.¹⁴

Two affected males and 2 obligate carriers in the 3-generation family were assessed. The clinical data are summarized in Table 1.

The propositus (III:1) was an 18-year-old man. He noticed night blindness and blurred vision at age 13 years but did not go to the hospital then. He visited our university hospital for examination when he was aged 18 years. Ophthalmologic examinations revealed pigmentation in midperipheral fundi (Figure 2A). On his most recent visit (when he was aged 19 years), Goldmann perimetry showed paracentral scotoma (isopter II-4e and III-4e) and depression of threshold values (isopter II-4e, 15°; III-4e, 30°; and V-4e, 45°). Both maximum-flash ERG and 30-Hz flicker ERG showed extinguished responses (Figure 2A). The propositus's mother (II:3), a heterozygous woman aged 39 years, did not have night blindness. She had high myopia with best-corrected visual acuity of 0.8 OU, and the fundus examination showed chorioretinal thinning, myopic optic discs, and peripapillary atrophy, which were consistent with high myopia. Few spicule formations were observed in the peripheral fundi (Figure 2B). The ERG showed a reduced response (Figure 2B). The propositus's grandfather (I:3), aged 66 years, had night blindness and severe myopia from childhood. Phacectomy was performed on one of his eyes at age 18 years and on the other eye at age 66 years because of high myopia. He had very poor visual acuity and severe myopic fundi with extensive chorioretinal atrophy as well as midperipheral pigmentation (Figure 2C). The ERG showed a distinguished response (Figure 2C). Goldmann perimetry results were unrecordable because of the low vision. The propositus's younger sister (III:2) noticed night blindness at age 17 years. Fundus examination showed bilateral chorioretinal atrophy that was consistent with high myopia (Figure 2D). The ERG showed a severely reduced response (Figure 2D) and the visual field was slightly reduced. The propositus's father and youngest sister had no symptoms or signs of RP, and their ERG and Goldmann perimetry results were normal.

Haplotype analysis showed that the affected individuals in the family shared a common haplotype with 5 markers (Figure 1). Direct polymerase chain reaction sequencing of affected individuals identified a 2-bp deletion at positions 1232 and 1233 (Figure 3) in exon ORF15, which was predicted to create an early termination at position 492 (Gly411fsTer492). The ORF15Gly411fsTer492 refers to a frameshift mutation in which Gly411 is the first amino acid altered and the termination of the ORF is at residue 492. The mutation was cosegregated with affected individuals in the family and was not observed in any of the unaffected family members or the group of normal controls.

RPGR accounts for up to 20% of all cases of RP,¹⁶ which is higher than any other RP locus. Mutations of RPGR are responsible for nonsystematic or systematic XLRP, XLCORD, X-linked cone dystrophy, and X-linked recessive atrophic macular degeneration. To date, at least 97 different mutations in exon ORF15 have been reported.¹⁷ However, the precise role of ORF15 is still unclear. Studying the genotype-phenotype correlation can elucidate the expressivity and penetration of the phenotype in the patient with a specific mutation in RPGR. By evaluating 2 different dog strains with mutations in different locations of ORF15, Zhang et al¹⁸ found that mutations toward the 3′ end of ORF15 and with subsequent shorter abnormal amino acid sequences produce higher retention of rod function and hence milder RP phenotypes. Sharon et al¹⁰ also proposed a hypothesis that mutations downstream of codon 445 in ORF15 strongly implicate the early preferential loss of cone function with slight-to-moderate loss of rod function. Consistent with their findings, 4 of the ORF15 mutations identified in families with XLCORD or X-linked cone dystrophy were located downstream of codon 445 of ORF15.^6,7,10 A recent study reported 2 nonsense mutations (codons 365 and 392) located in the 3′ end of the ORF15 highly repetitive stretch, indicating that mutations toward the 3′ end (downstream of codon 365) of the highly repetitive region may cause cone-predominant dysfunction.⁹ It is noted that there are more XLRP mutations in the 3′ end of the highly repetitive stretch of ORF15 than XLCORD mutations, and the region downstream of codon 445 harbors more XLCORD mutations than XLRP mutations. Hypothetically setting codon 365 as a point of demarcation, there were at least 24 mutations downstream of codon 365 that were identified in patients diagnosed with XLRP. Even setting codon 445 as a point of demarcation, 10 different mutations downstream of codon 445 had been reported, and this most 3′ coding region still harbors 4 mutations for confirmed XLRP and 1 mutation responsible for both XLRP and XLCORD. We retrospectively summarized the mutations downstream of codon 365 (nucleotide position 1095) from the previous studies (Table 2). Of the 32 mutations, 7 were reported in 8 unrelated families with XLCORD and 1 mutation was reported in a family with X-linked recessive atrophic macular degeneration⁸ that was subsequently suspected by other investigators to have had XLCORD.⁹ In addition, a mutation (g.ORF15 + 1563-1566delAAGT) very close to the 3′ end of the ORF15 coding sequence was reported in a family with XLRP,²² but the investigators admitted that the patients had cone-rod dystrophy in their records. Most recently, Pelletier et al¹⁹ reported the mutation g.ORF15 + 1641_1642delAA closest to the 3′ end of ORF15 in a family with XLCORD. Interestingly, the mutation of ORF15 + 1339-1340delAG was detected in 3 families with XLCORD^6,10,20 as well as in 1 family with XLRP.²⁰ The remaining 24 mutations, including the mutation identified in our study, were reported in 32 unrelated families with XLRP.^{5-7,10,11,14,19-22} It is worthwhile to note that, to our knowledge, few detailed phenotypes were documented in these families with XLRP with mutations at the 3′ end of the coding sequence of exon ORF15.¹¹

The detailed phenotypes of the family in our study are consistent with XLRP. The affected males had night blindness at early ages and experienced progressive deterioration in central vision with no apparent macular degeneration in their fundi. The 2 male patients had typical RP symptoms and extinguished ERG responses (on both maximum-flash and 30-Hz flicker ERG), which were consistent with previously reported XLRP cases that had severe retinal degeneration with both rod and cone affected.^13,23 It is well known that female carriers have a wide spectrum of phenotypes ranging from normal to severe. In our study, the 2 carriers were highly myopic, a known characteristic of XLRP carriers. Both showed reduced ERG responses, and Goldmann perimetry testing showed reduction in individual III:2 (the data for II:3 were not available). It appears that the patients and carriers with this mutation have both rod and cone dysfunctions, which is in contrast to the previous hypothesis proposed by Sharon et al.¹⁰ Because the numbers of families with XLCORD or X-linked cone dystrophy were small and a limited number of families with XLRP with mutations in the 3′ end of the exon ORF15 coding sequence were described, further studies are required to elucidate the significance of the mutation location and distribution. Both RP and cone-rod dystrophy are very heterogeneous in clinical symptoms. Because some cases of cone-rod dystrophy may have only minor macular retinal pigment epithelium atrophy without typical bull's-eye maculopathy and may show both cone and rod involvements in ERGs, which also appear in some RP cases (especially in XLRP with rapid deterioration and severe degeneration of both rods and cones), the similarity in symptoms leads to an indecisive diagnosis. More definite clinical diagnosis and classification are necessary for a differential diagnosis of XLCORD and for elucidating the role of ORF15 in cone and rod involvements.

We have identified a novel mutation in a family with XLRP and have documented the clinical manifestations. The clinical findings are consistent with previous reports of XLRP phenotypes associated with mutations in RPGR, leading us to speculate that the mutation 3′ toward the highly repetitive stretch of ORF15 or in the 3′ end of the ORF15 coding sequence may be the cause of XLRP as well as XLCORD, and the mutation in the 3′ end of ORF15 can lead to a spectrum of phenotypes. Our findings are consistent with the previous studies listed in Table 2 that have suggested the tendency of XLRP mutations to cluster in the highly repetitive region and support the observation that RPGR exon ORF15 mutations (including those that are located at the 3′ end of the exon ORF15 coding sequence) can cause different phenotypes. The summarized ORF15 mutations downstream from codon 365 (Table 2) also clearly suggest that the cone-predominant phenotype-related mutations can be located irregularly in the exon ORF15 coding sequence, and the number of families is not sufficient for a meaningful conclusion given the rare occurrence of this phenotype. The detailed phenotypes of patients with the mutation close to the 3′ end of the ORF15 coding sequence may be helpful in comparing the XLRP with the increasing XLCORD caused by ORF15 mutations and in establishing the significance of the mutation distribution and genotype-phenotype correlation. Further studies exploring the correspondence between mutations and phenotypes are required to gain insight into the pathogenesis of XLRP and XLCORD.

Correspondence: Nobuhisa Nao-i, MD, Department of Ophthalmology & Visual Science, Faculty of Medicine, University of Miyazaki, Kihara 5200, Kiyotake, Miyazaki 889-1692, Japan (naoi@fc.miyazaki-u.ac.jp).

Submitted for Publication: September 26, 2006; final revision received February 9, 2007; accepted February 28, 2007.

Financial Disclosure: None reported.

Funding/Support: This work was supported in part by Grant-in-Aid for Scientific Research (C) 17591843 from the Japan Society for the Promotion of Science.